初始化项目，由ModelHub XC社区提供模型
Model: OpenBMB/MiniCPM5-2B-Base Source: Original Platform
This commit is contained in:
ModelHub XC
48
.gitattributes
vendored
Normal file
48
.gitattributes
vendored
Normal file
@@ -0,0 +1,48 @@
|
||||
*.7z filter=lfs diff=lfs merge=lfs -text
|
||||
*.arrow filter=lfs diff=lfs merge=lfs -text
|
||||
*.bin filter=lfs diff=lfs merge=lfs -text
|
||||
*.bin.* filter=lfs diff=lfs merge=lfs -text
|
||||
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
||||
*.ftz filter=lfs diff=lfs merge=lfs -text
|
||||
*.gz filter=lfs diff=lfs merge=lfs -text
|
||||
*.h5 filter=lfs diff=lfs merge=lfs -text
|
||||
*.joblib filter=lfs diff=lfs merge=lfs -text
|
||||
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
||||
*.model filter=lfs diff=lfs merge=lfs -text
|
||||
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
||||
*.onnx filter=lfs diff=lfs merge=lfs -text
|
||||
*.ot filter=lfs diff=lfs merge=lfs -text
|
||||
*.parquet filter=lfs diff=lfs merge=lfs -text
|
||||
*.pb filter=lfs diff=lfs merge=lfs -text
|
||||
*.pt filter=lfs diff=lfs merge=lfs -text
|
||||
*.pth filter=lfs diff=lfs merge=lfs -text
|
||||
*.rar filter=lfs diff=lfs merge=lfs -text
|
||||
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
||||
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
||||
*.tflite filter=lfs diff=lfs merge=lfs -text
|
||||
*.tgz filter=lfs diff=lfs merge=lfs -text
|
||||
*.xz filter=lfs diff=lfs merge=lfs -text
|
||||
*.zip filter=lfs diff=lfs merge=lfs -text
|
||||
*.zstandard filter=lfs diff=lfs merge=lfs -text
|
||||
*.tfevents* filter=lfs diff=lfs merge=lfs -text
|
||||
*.db* filter=lfs diff=lfs merge=lfs -text
|
||||
*.ark* filter=lfs diff=lfs merge=lfs -text
|
||||
**/*ckpt*data* filter=lfs diff=lfs merge=lfs -text
|
||||
**/*ckpt*.meta filter=lfs diff=lfs merge=lfs -text
|
||||
**/*ckpt*.index filter=lfs diff=lfs merge=lfs -text
|
||||
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
||||
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
||||
*.gguf* filter=lfs diff=lfs merge=lfs -text
|
||||
*.ggml filter=lfs diff=lfs merge=lfs -text
|
||||
*.llamafile* filter=lfs diff=lfs merge=lfs -text
|
||||
*.pt2 filter=lfs diff=lfs merge=lfs -text
|
||||
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
||||
*.npy filter=lfs diff=lfs merge=lfs -text
|
||||
*.npz filter=lfs diff=lfs merge=lfs -text
|
||||
*.pickle filter=lfs diff=lfs merge=lfs -text
|
||||
*.pkl filter=lfs diff=lfs merge=lfs -text
|
||||
*.tar filter=lfs diff=lfs merge=lfs -text
|
||||
*.wasm filter=lfs diff=lfs merge=lfs -text
|
||||
*.zst filter=lfs diff=lfs merge=lfs -text
|
||||
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
||||
"/tokenizer.json" filter=lfs diff=lfs merge=lfs -text
|
||||
533
README-cn.md
Normal file
533
README-cn.md
Normal file
@@ -0,0 +1,533 @@
|
||||
---
|
||||
license: apache-2.0
|
||||
language:
|
||||
- zh
|
||||
- en
|
||||
library_name: transformers
|
||||
pipeline_tag: text-generation
|
||||
tags:
|
||||
- minicpm
|
||||
- minicpm5
|
||||
- llama
|
||||
- text-generation
|
||||
- long-context
|
||||
- tool-calling
|
||||
- on-device
|
||||
- edge-ai
|
||||
datasets:
|
||||
- openbmb/Ultra-FineWeb
|
||||
- openbmb/UltraX-Preview
|
||||
- openbmb/Ultra-FineWeb-L3
|
||||
- openbmb/UltraData-Math
|
||||
- openbmb/UltraData-Code
|
||||
- openbmb/UltraData-SFT-2605
|
||||
- openbmb/UltraData-SFT-Agent-2609
|
||||
- openbmb/UltraData-RL-2609
|
||||
---
|
||||
|
||||
|
||||
|
||||
[MiniCPM 技术报告](https://arxiv.org/pdf/2506.07900) | [MiniCPM 知识库](https://modelbest.feishu.cn/wiki/UtWxwcERfiRIpIkBOjuc3h9tn1D) | [GitHub 仓库](https://github.com/OpenBMB/MiniCPM) | [UltraData](https://ultradata.openbmb.cn/) | [在线 Demo](https://huggingface.co/spaces/openbmb/MiniCPM5-2B-Demo)
|
||||
|
||||
[English](https://huggingface.co/openbmb/MiniCPM5-2B/blob/main/README.md) | 中文
|
||||
|
||||
## 亮点
|
||||
|
||||
我们正式发布 **MiniCPM5-2B**，这是继 [MiniCPM5-1B](https://huggingface.co/openbmb/MiniCPM5-1B) 之后 **MiniCPM5** 系列的第二个模型。它是一款面向端侧、本地部署和资源受限场景的 2B 稠密 Transformer，能够达到同尺寸开源模型 SOTA 水平。
|
||||
|
||||
🏆 **同尺寸开源模型 SOTA**：与同尺寸优秀开源模型相比，MiniCPM5-2B 在该对比范围内达到 SOTA 水平，整体表现可与 4B 级模型竞争，并在代码、数学、长文本、工具调用和 Agent 任务上展现出明显优势。
|
||||
|
||||
<div id="capability-comparison-radar" class="radar-visual" role="img" aria-label="能力雷达图：对比 MiniCPM5-2B 与 4B 级模型；每个维度分别将最高分归一化为 100%。">
|
||||
<style>
|
||||
#capability-comparison-radar {
|
||||
--foreground: #171717;
|
||||
display: block; width: 100%; max-width: 620px; margin: 0 auto; background: #fff;
|
||||
font-family: -apple-system, BlinkMacSystemFont, "Segoe UI", Roboto, "PingFang SC", "Hiragino Sans GB", "Microsoft YaHei", sans-serif;
|
||||
}
|
||||
#capability-comparison-radar svg { display: block; width: 100%; height: auto; background: #fff; }
|
||||
#capability-comparison-radar .title { fill: var(--foreground); font-size: 16px; font-weight: 600; letter-spacing: 0; }
|
||||
#capability-comparison-radar .axis-label { fill: #333; font-size: 12px; font-weight: 600; }
|
||||
#capability-comparison-radar .ring { fill: none; stroke: rgba(128,128,128,.18); stroke-width: .8; }
|
||||
#capability-comparison-radar .spoke { stroke: rgba(128,128,128,.25); stroke-width: .8; }
|
||||
#capability-comparison-radar .ring-label { fill: #8a8a8a; font-size: 9px; }
|
||||
#capability-comparison-radar .series { stroke-linejoin: round; }
|
||||
#capability-comparison-radar .series.primary { stroke-width: 2.4; }
|
||||
#capability-comparison-radar .legend-label { fill: #171717; font-size: 12px; font-weight: 600; }
|
||||
#capability-comparison-radar .legend-average { fill: #666; font-size: 11px; }
|
||||
#capability-comparison-radar .legend-swatch { rx: 3; }
|
||||
@media (max-width: 900px) { #capability-comparison-radar { overflow-x: auto; } #capability-comparison-radar svg { min-width: 620px; } }
|
||||
</style>
|
||||
<svg viewBox="0 0 680 560" aria-hidden="true">
|
||||
<g transform="translate(-20 0)">
|
||||
<text x="40" y="28" class="title" text-anchor="start">各能力维度雷达图</text>
|
||||
<polygon points="340.0,235.0 362.5,243.2 374.5,263.9 370.3,287.5 352.0,302.9 328.0,302.9 309.7,287.5 305.5,263.9 317.5,243.2" class="ring" stroke="rgba(128,128,128,.18)" stroke-width=".8"/>
|
||||
<text x="344.0" y="238.0" class="ring-label">20%</text>
|
||||
<polygon points="340.0,200.0 385.0,216.4 408.9,257.8 400.6,305.0 363.9,335.8 316.1,335.8 279.4,305.0 271.1,257.8 295.0,216.4" class="ring" stroke="rgba(128,128,128,.18)" stroke-width=".8"/>
|
||||
<text x="344.0" y="203.0" class="ring-label">40%</text>
|
||||
<polygon points="340.0,165.0 407.5,189.6 443.4,251.8 430.9,322.5 375.9,368.7 304.1,368.7 249.1,322.5 236.6,251.8 272.5,189.6" class="ring" stroke="rgba(128,128,128,.18)" stroke-width=".8"/>
|
||||
<text x="344.0" y="168.0" class="ring-label">60%</text>
|
||||
<polygon points="340.0,130.0 430.0,162.8 477.9,245.7 461.2,340.0 387.9,401.6 292.1,401.6 218.8,340.0 202.1,245.7 250.0,162.8" class="ring" stroke="rgba(128,128,128,.18)" stroke-width=".8"/>
|
||||
<text x="344.0" y="133.0" class="ring-label">80%</text>
|
||||
<polygon points="340.0,95.0 452.5,135.9 512.3,239.6 491.6,357.5 399.9,434.4 280.1,434.4 188.4,357.5 167.7,239.6 227.5,135.9" class="ring" stroke="rgba(29,111,208,.28)" stroke-width="1"/>
|
||||
<text x="344.0" y="98.0" class="ring-label">100%</text>
|
||||
<line x1="340" y1="270" x2="340.0" y2="95.0" class="spoke"/>
|
||||
<text x="340.0" y="63.5" class="axis-label" text-anchor="middle">代码推理</text>
|
||||
<line x1="340" y1="270" x2="452.5" y2="135.9" class="spoke"/>
|
||||
<text x="472.7" y="111.8" class="axis-label" text-anchor="start">数学推理</text>
|
||||
<line x1="340" y1="270" x2="512.3" y2="239.6" class="spoke"/>
|
||||
<text x="543.4" y="234.1" class="axis-label" text-anchor="start">指令遵循</text>
|
||||
<line x1="340" y1="270" x2="491.6" y2="357.5" class="spoke"/>
|
||||
<text x="518.8" y="373.2" class="axis-label" text-anchor="start">综合知识</text>
|
||||
<line x1="340" y1="270" x2="399.9" y2="434.4" class="spoke"/>
|
||||
<text x="410.6" y="464.0" class="axis-label" text-anchor="start">长文本</text>
|
||||
<line x1="340" y1="270" x2="280.1" y2="434.4" class="spoke"/>
|
||||
<text x="269.4" y="464.0" class="axis-label" text-anchor="end">工具调用</text>
|
||||
<line x1="340" y1="270" x2="188.4" y2="357.5" class="spoke"/>
|
||||
<text x="161.2" y="373.3" class="axis-label" text-anchor="end">代码智能体</text>
|
||||
<line x1="340" y1="270" x2="167.7" y2="239.6" class="spoke"/>
|
||||
<text x="136.6" y="234.1" class="axis-label" text-anchor="end">搜索智能体</text>
|
||||
<line x1="340" y1="270" x2="227.5" y2="135.9" class="spoke"/>
|
||||
<text x="207.3" y="111.8" class="axis-label" text-anchor="end">通用智能体</text>
|
||||
<polygon points="340.0,136.7 450.1,138.8 498.4,242.1 491.6,357.5 398.4,430.3 289.5,408.8 188.4,357.5 188.2,243.2 249.3,161.9" class="series" fill="#E85D4C" fill-opacity=".10" stroke="#E85D4C" stroke-width="1.5"/>
|
||||
<circle cx="340.0" cy="136.7" r="3" fill="#E85D4C"/>
|
||||
<circle cx="450.1" cy="138.8" r="3" fill="#E85D4C"/>
|
||||
<circle cx="498.4" cy="242.1" r="3" fill="#E85D4C"/>
|
||||
<circle cx="491.6" cy="357.5" r="3" fill="#E85D4C"/>
|
||||
<circle cx="398.4" cy="430.3" r="3" fill="#E85D4C"/>
|
||||
<circle cx="289.5" cy="408.8" r="3" fill="#E85D4C"/>
|
||||
<circle cx="188.4" cy="357.5" r="3" fill="#E85D4C"/>
|
||||
<circle cx="188.2" cy="243.2" r="3" fill="#E85D4C"/>
|
||||
<circle cx="249.3" cy="161.9" r="3" fill="#E85D4C"/>
|
||||
<polygon points="340.0,134.4 448.9,140.2 512.3,239.6 462.2,340.5 367.8,346.5 308.0,358.0 231.0,332.9 242.3,252.8 250.1,162.9" class="series" fill="#2A9D8F" fill-opacity=".10" stroke="#2A9D8F" stroke-width="1.5"/>
|
||||
<circle cx="340.0" cy="134.4" r="3" fill="#2A9D8F"/>
|
||||
<circle cx="448.9" cy="140.2" r="3" fill="#2A9D8F"/>
|
||||
<circle cx="512.3" cy="239.6" r="3" fill="#2A9D8F"/>
|
||||
<circle cx="462.2" cy="340.5" r="3" fill="#2A9D8F"/>
|
||||
<circle cx="367.8" cy="346.5" r="3" fill="#2A9D8F"/>
|
||||
<circle cx="308.0" cy="358.0" r="3" fill="#2A9D8F"/>
|
||||
<circle cx="231.0" cy="332.9" r="3" fill="#2A9D8F"/>
|
||||
<circle cx="242.3" cy="252.8" r="3" fill="#2A9D8F"/>
|
||||
<circle cx="250.1" cy="162.9" r="3" fill="#2A9D8F"/>
|
||||
<polygon points="340.0,189.3 411.4,184.9 502.9,241.3 455.3,336.6 356.7,315.9 288.5,411.4 320.8,281.1 266.8,257.1 298.8,220.9" class="series" fill="#E09F3E" fill-opacity=".10" stroke="#E09F3E" stroke-width="1.5"/>
|
||||
<circle cx="340.0" cy="189.3" r="3" fill="#E09F3E"/>
|
||||
<circle cx="411.4" cy="184.9" r="3" fill="#E09F3E"/>
|
||||
<circle cx="502.9" cy="241.3" r="3" fill="#E09F3E"/>
|
||||
<circle cx="455.3" cy="336.6" r="3" fill="#E09F3E"/>
|
||||
<circle cx="356.7" cy="315.9" r="3" fill="#E09F3E"/>
|
||||
<circle cx="288.5" cy="411.4" r="3" fill="#E09F3E"/>
|
||||
<circle cx="320.8" cy="281.1" r="3" fill="#E09F3E"/>
|
||||
<circle cx="266.8" cy="257.1" r="3" fill="#E09F3E"/>
|
||||
<circle cx="298.8" cy="220.9" r="3" fill="#E09F3E"/>
|
||||
<polygon points="340.0,95.0 452.5,135.9 499.8,241.8 476.2,348.7 399.9,434.4 280.1,434.4 220.0,339.3 167.7,239.6 227.5,135.9" class="series primary" fill="#1D6FD0" fill-opacity=".22" stroke="#1D6FD0" stroke-width="2.4"/>
|
||||
<circle cx="340.0" cy="95.0" r="3" fill="#1D6FD0"/>
|
||||
<circle cx="452.5" cy="135.9" r="3" fill="#1D6FD0"/>
|
||||
<circle cx="499.8" cy="241.8" r="3" fill="#1D6FD0"/>
|
||||
<circle cx="476.2" cy="348.7" r="3" fill="#1D6FD0"/>
|
||||
<circle cx="399.9" cy="434.4" r="3" fill="#1D6FD0"/>
|
||||
<circle cx="280.1" cy="434.4" r="3" fill="#1D6FD0"/>
|
||||
<circle cx="220.0" cy="339.3" r="3" fill="#1D6FD0"/>
|
||||
<circle cx="167.7" cy="239.6" r="3" fill="#1D6FD0"/>
|
||||
<circle cx="227.5" cy="135.9" r="3" fill="#1D6FD0"/>
|
||||
<rect x="40" y="490" width="18" height="18" class="legend-swatch" fill="#1D6FD0"/>
|
||||
<text x="66" y="504" class="legend-label">MiniCPM5-2B</text>
|
||||
<text x="66" y="520" class="legend-average">平均分 53.9</text>
|
||||
<rect x="195" y="490" width="18" height="18" class="legend-swatch" fill="#E85D4C"/>
|
||||
<text x="221" y="504" class="legend-label">Qwen3.5-4B</text>
|
||||
<text x="221" y="520" class="legend-average">平均分 51.1</text>
|
||||
<rect x="350" y="490" width="18" height="18" class="legend-swatch" fill="#2A9D8F"/>
|
||||
<text x="376" y="504" class="legend-label">granite-4.2-3B</text>
|
||||
<text x="376" y="520" class="legend-average">平均分 42.7</text>
|
||||
<rect x="505" y="490" width="18" height="18" class="legend-swatch" fill="#E09F3E"/>
|
||||
<text x="531" y="504" class="legend-label">LFM2.5-2.6B</text>
|
||||
<text x="531" y="520" class="legend-average">平均分 33.2</text>
|
||||
<text x="340" y="548" class="legend-average" text-anchor="middle">每条轴：该维最高分 = 100%</text>
|
||||
</g>
|
||||
</svg>
|
||||
</div>
|
||||
|
||||
📂 **开放高质量数据**：与模型一同开源其背后的高质量训练数据，均属于 [UltraData](https://ultradata.openbmb.cn/) 数据体系：[UltraX](https://huggingface.co/datasets/openbmb/UltraX-Preview)，高质量网页预训练数据集；[UltraData-Code](https://huggingface.co/datasets/openbmb/UltraData-Code)，L0–L3 分级代码治理，推动代码能力显著跃升；[UltraData-SFT-Agent-2609](https://huggingface.co/datasets/openbmb/UltraData-SFT-Agent-2609)，50 万 Agent 训练样本，赋能端侧 Agent 综合能力提升；[UltraData-RL-2609](https://huggingface.co/datasets/openbmb/UltraData-RL-2609)，超 8 万条高质量 RL 训练样本，覆盖数学、代码、通用知识与长文本推理。
|
||||
|
||||
## 模型列表
|
||||
|
||||
你可以按运行环境选择对应模型格式：
|
||||
|
||||
**MiniCPM5-2B**
|
||||
|
||||
- **[MiniCPM5-2B](https://huggingface.co/openbmb/MiniCPM5-2B)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-2B) · BF16 正式版（经 RL + OPD 后训练） **👈 当前页面**
|
||||
- **[MiniCPM5-2B-SFT](https://huggingface.co/openbmb/MiniCPM5-2B-SFT)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-2B-SFT) · BF16 SFT 单独 checkpoint（RL / OPD 之前）
|
||||
- **[MiniCPM5-2B-Midtrain](https://huggingface.co/openbmb/MiniCPM5-2B-Midtrain)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-2B-Midtrain) · BF16 mid-training checkpoint（SFT 之前）
|
||||
- **[MiniCPM5-2B-Base](https://huggingface.co/openbmb/MiniCPM5-2B-Base)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-2B-Base) · BF16 base checkpoint（仅预训练） **👈 当前页面**
|
||||
- **[MiniCPM5-2B-GGUF](https://huggingface.co/openbmb/MiniCPM5-2B-GGUF)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-2B-GGUF) · GGUF，适用于 llama.cpp / Ollama / LM Studio
|
||||
- **[MiniCPM5-2B-MLX](https://huggingface.co/openbmb/MiniCPM5-2B-MLX)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-2B-MLX) · MLX / 4bit，适用于 Apple Silicon
|
||||
- **[MiniCPM5-2B-GPTQ](https://huggingface.co/openbmb/MiniCPM5-2B-GPTQ)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-2B-GPTQ) · GPTQ / 4bit 量化模型
|
||||
- **[MiniCPM5-2B-DSpark](https://huggingface.co/openbmb/MiniCPM5-2B-DSpark)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-2B-DSpark) · DSpark 草稿模型，用于推理加速
|
||||
- **[MiniCPM5-2B-DSpark-GGUF](https://huggingface.co/openbmb/MiniCPM5-2B-DSpark-GGUF)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-2B-DSpark-GGUF) · GGUF 版本的 DSpark 草稿模型
|
||||
- **[MiniCPM5-2B-LiteRT](https://huggingface.co/litert-community/MiniCPM5-2B)** · [ModelScope](https://www.modelscope.cn/models/litert-community/MiniCPM5-2B) · MiniCPM5-2B 的 LiteRT-LM 版本
|
||||
|
||||
**MiniCPM5-1B**
|
||||
|
||||
- **[MiniCPM5-1B](https://huggingface.co/openbmb/MiniCPM5-1B)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-1B) · BF16 正式版（经 RL + OPD 后训练）
|
||||
- **[MiniCPM5-1B-SFT](https://huggingface.co/openbmb/MiniCPM5-1B-SFT)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-1B-SFT) · BF16 SFT 单独 checkpoint（RL / OPD 之前）
|
||||
- **[MiniCPM5-1B-Base](https://huggingface.co/openbmb/MiniCPM5-1B-Base)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-1B-Base) · BF16 base checkpoint（仅预训练）
|
||||
- **[MiniCPM5-1B-GGUF](https://huggingface.co/openbmb/MiniCPM5-1B-GGUF)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-1B-GGUF) · GGUF，适用于 llama.cpp / Ollama / LM Studio
|
||||
- **[MiniCPM5-1B-MLX](https://huggingface.co/openbmb/MiniCPM5-1B-MLX)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-1B-MLX) · MLX / 4bit，适用于 Apple Silicon
|
||||
|
||||
## 模型信息
|
||||
|
||||
MiniCPM5-2B 具有以下特性：
|
||||
|
||||
- **类型**：Causal Language Model
|
||||
- **架构**：标准 `LlamaForCausalLM`
|
||||
- **参数数量**：2,516,756,480
|
||||
- **非嵌入参数数量**：1,981,982,720
|
||||
- **层数**：42
|
||||
- **注意力头（GQA）**：16 个 Q heads / 2 个 KV heads
|
||||
- **上下文长度**：131,072
|
||||
|
||||
## 简介
|
||||
|
||||
MiniCPM5-2B 是 MiniCPM5 系列的第二个模型，面向本地助手、coding agent、工具调用流程以及需要紧凑模型的推理场景。它在较小部署成本下提供原生长上下文能力。
|
||||
|
||||
## 评测结果
|
||||
|
||||
我们选取 **LFM2.5-2.6B**、**Qwen3.5-2B**、**Gemma-4-E2B-it** 等同尺寸开源模型进行横向比较，并同时列出 **Qwen3.5-4B**、**granite-4.2-3B**、**Nemotron-3-Nano-4B**、**Gemma-4-E4B-it**、**LFM2.5-8B-A1B** 等更大规模模型作为参考。
|
||||
|
||||
在这组对比中，MiniCPM5-2B 达到同尺寸开源模型 SOTA 水平（平均分 53.9 ），也超过了参与对比的全部更大规模模型（最高 51.1）。其优势主要体现在代码推理、数学推理、长文本、工具调用与多个智能体任务上。
|
||||
|
||||
<div style="width:100%;max-width:1080px;margin:0 auto;padding:16px 0;background:#fff;
|
||||
font-family:-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,'PingFang SC',
|
||||
'Hiragino Sans GB','Microsoft YaHei',sans-serif;color:#171717">
|
||||
<h1 style="margin:0 0 14px;font-size:22px;font-weight:700;color:#1D6FD0;
|
||||
letter-spacing:0.02em">MiniCPM5-2B 与基线模型评测结果</h1>
|
||||
<table class="vl-table" style="width:100%;margin:0;table-layout:fixed;border-collapse:collapse;font-size:13px;font-variant-numeric:tabular-nums"><thead><tr><th rowspan="2" style="padding:7px 5px;text-align:left;font-weight:600;border-bottom:2px solid #1D6FD0;color:#1D6FD0;width:18%"></th><th rowspan="2" style="padding:7px 4px;text-align:center;font-weight:600;border-bottom:2px solid #1D6FD0;color:#1D6FD0;font-size:12px;width:9.111%;background:rgba(29, 111, 208, 0.08);vertical-align:middle;word-break:normal;">MiniCPM5-2B</th><th colspan="3" style="padding:6px 4px;text-align:center;font-weight:600;color:#1D6FD0;font-size:13px;border-bottom:1px solid rgba(29, 111, 208, 0.2);border-left:1px solid rgba(29, 111, 208, 0.25);">2B 级模型</th><th colspan="5" style="padding:6px 4px;text-align:center;font-weight:600;color:#1D6FD0;font-size:13px;border-bottom:1px solid rgba(29, 111, 208, 0.2);border-left:1px solid rgba(29, 111, 208, 0.25);">4B 级模型</th></tr><tr><th style="padding:6px 4px;text-align:center;font-weight:500;border-bottom:2px solid #1D6FD0;color:#1D6FD0;font-size:11px;width:9.111%;border-left:1px solid rgba(29, 111, 208, 0.25);word-break:normal;vertical-align:middle;">LFM2.5-2.6B</th><th style="padding:6px 4px;text-align:center;font-weight:500;border-bottom:2px solid #1D6FD0;color:#1D6FD0;font-size:11px;width:9.111%;word-break:normal;vertical-align:middle;">Qwen3.5-2B</th><th style="padding:6px 4px;text-align:center;font-weight:500;border-bottom:2px solid #1D6FD0;color:#1D6FD0;font-size:11px;width:9.111%;word-break:normal;vertical-align:middle;">Gemma-4-E2B-it</th><th style="padding:6px 4px;text-align:center;font-weight:500;border-bottom:2px solid #1D6FD0;color:#1D6FD0;font-size:11px;width:9.111%;border-left:1px solid rgba(29, 111, 208, 0.25);word-break:normal;vertical-align:middle;">Qwen3.5-4B</th><th style="padding:6px 4px;text-align:center;font-weight:500;border-bottom:2px solid #1D6FD0;color:#1D6FD0;font-size:11px;width:9.111%;word-break:normal;vertical-align:middle;">granite-4.2-3B</th><th style="padding:6px 4px;text-align:center;font-weight:500;border-bottom:2px solid #1D6FD0;color:#1D6FD0;font-size:11px;width:9.111%;word-break:normal;vertical-align:middle;">Nemotron-3-Nano-4B</th><th style="padding:6px 4px;text-align:center;font-weight:500;border-bottom:2px solid #1D6FD0;color:#1D6FD0;font-size:11px;width:9.111%;word-break:normal;vertical-align:middle;">Gemma-4-E4B-it</th><th style="padding:6px 4px;text-align:center;font-weight:500;border-bottom:2px solid #1D6FD0;color:#1D6FD0;font-size:11px;width:9.111%;word-break:normal;vertical-align:middle;">LFM2.5-8B-A1B</th></tr></thead><tbody>
|
||||
<tr style="background:rgba(29, 111, 208, 0.03)"><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">平均分</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;font-weight:600;"><strong style="color:#1D6FD0">53.9</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;font-weight:600;">33.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;font-weight:600;">28.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;font-weight:600;">24.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;font-weight:600;">51.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;font-weight:600;">42.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;font-weight:600;">32.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;font-weight:600;">31.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;font-weight:600;">28.4</td></tr>
|
||||
<tr><td colspan="10" style="padding:5px 10px;font-weight:600;color:#1D6FD0;border-bottom:1px solid rgba(29, 111, 208, 0.2);background:rgba(29, 111, 208, 0.14)">代码推理</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">LiveCodeBench v6</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">69.1</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">42.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">20.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">42.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">56.4</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">58.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">50.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">53.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">39.8</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">LCB-Pro 25Q2 (Easy)</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">68.0</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">30.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">10.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">27.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">58.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">54.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">51.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">45.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">27.8</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">LCB-Pro 25Q2 (Medium)</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">17.5</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">0.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">7.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">5.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">5.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">1.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.0</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">OJBench</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">32.5</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">11.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">2.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">11.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">24.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">21.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">20.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">19.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">8.2</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">SciCode (wbg)</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">26.3</strong><sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">14.2<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">2.8<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">20.9<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">16.1<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">24.9<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">16.4<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">24.4<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">7.8<sup style="font-size:0.72em;opacity:0.7">†</sup></td></tr>
|
||||
<tr><td colspan="10" style="padding:5px 10px;font-weight:600;color:#1D6FD0;border-bottom:1px solid rgba(29, 111, 208, 0.2);background:rgba(29, 111, 208, 0.14)">数学推理</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">AIME 2025</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">86.5</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">41.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">29.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">31.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">78.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">79.4</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">56.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">37.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">46.0</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">AIME 2026</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">86.5</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">45.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">29.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">39.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">82.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">83.5</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">62.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">45.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">56.7</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">HMMT Feb 2026</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>63.8</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">33.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">20.5</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">17.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">64.0</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">60.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">51.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">30.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">38.5</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">MATH-500</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>94.6</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">89.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">85.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">85.4</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">99.0</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">97.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">91.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">88.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">93.2</td></tr>
|
||||
<tr><td colspan="10" style="padding:5px 10px;font-weight:600;color:#1D6FD0;border-bottom:1px solid rgba(29, 111, 208, 0.2);background:rgba(29, 111, 208, 0.14)">指令遵循</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">IFBench</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>66.3</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">59.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">46.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">25.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">59.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">73.0</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">58.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">28.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">51.0</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">IFEval</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;">86.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>93.4</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">77.5</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">31.4</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">90.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">93.7</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">88.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">44.4</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">90.8</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">Multi-IF</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;">71.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">76.8</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">57.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">40.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">73.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">75.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">65.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">45.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">71.4</td></tr>
|
||||
<tr><td colspan="10" style="padding:5px 10px;font-weight:600;color:#1D6FD0;border-bottom:1px solid rgba(29, 111, 208, 0.2);background:rgba(29, 111, 208, 0.14)">综合知识</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">MMLU-Pro</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>70.8</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">65.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">64.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">56.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">78.0</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">65.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">65.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">68.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">63.1</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">MMLU-Redux</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>84.7</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">80.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">80.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">71.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">88.7</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">78.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">79.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">83.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">80.0</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">HLE</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>8.9</strong><sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">6.2<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">2.6<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">4.8<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">9.9</strong><sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">6.6<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">4.9<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">3.8<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">6.9<sup style="font-size:0.72em;opacity:0.7">†</sup></td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">GPQA-Diamond</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>70.2</strong><sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">55.8<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">45.6<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">43.3<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">77.1</strong><sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">55.9<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">51.3<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">57.6<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">51.3<sup style="font-size:0.72em;opacity:0.7">†</sup></td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">SuperGPQA</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>40.8</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">26.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">38.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">30.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">52.8</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">39.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">37.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">38.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">34.5</td></tr>
|
||||
<tr><td colspan="10" style="padding:5px 10px;font-weight:600;color:#1D6FD0;border-bottom:1px solid rgba(29, 111, 208, 0.2);background:rgba(29, 111, 208, 0.14)">长文本</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">AA-LCR</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>59.0</strong><sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">5.3<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">28.7<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">17.0<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">61.0</strong><sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">24.3<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">17.3<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">33.0<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.0<sup style="font-size:0.72em;opacity:0.7">†</sup></td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">NoLiMa</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">68.1</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">0.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">17.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">3.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">43.5</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">5.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">1.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">2.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.5</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">LongBenchPro</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>44.8</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">23.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">8.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">42.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">58.4</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">34.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">27.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">53.5</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">19.6</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">LongBench v2</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>43.7</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">30.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">24.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">33.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">47.3</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">36.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">32.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">42.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">30.4</td></tr>
|
||||
<tr><td colspan="10" style="padding:5px 10px;font-weight:600;color:#1D6FD0;border-bottom:1px solid rgba(29, 111, 208, 0.2);background:rgba(29, 111, 208, 0.14)">工具调用</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">τ³-Bench Banking</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">20.8</strong><sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">7.2<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">2.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">3.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">6.8<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">5.6<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">1.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">4.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">3.4</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">τ²-Bench Telecom</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">97.1</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">90.4</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">69.0<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">20.8<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">92.1<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">40.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">28.1<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">20.8<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">16.1<sup style="font-size:0.72em;opacity:0.7">†</sup></td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">BFCL v4</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">66.6</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">61.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">43.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">36.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">56.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">52.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">43.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">47.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">49.2</td></tr>
|
||||
<tr><td colspan="10" style="padding:5px 10px;font-weight:600;color:#1D6FD0;border-bottom:1px solid rgba(29, 111, 208, 0.2);background:rgba(29, 111, 208, 0.14)">代码智能体</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">SWE-bench Verified</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">46.4</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">6.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">5.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">2.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">33.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">36.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">3.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">15.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.4</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">SWE-bench Pro</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>14.4</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">0.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">28.2</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">12.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">3.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.4</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">Terminal-Bench v2.1</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>8.6</strong><sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">4.5<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">3.0<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.4<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">25.8</strong><sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">13.9<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">3.8<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">1.9<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">1.9</td></tr>
|
||||
<tr><td colspan="10" style="padding:5px 10px;font-weight:600;color:#1D6FD0;border-bottom:1px solid rgba(29, 111, 208, 0.2);background:rgba(29, 111, 208, 0.14)">搜索智能体</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">BrowseComp-ZH</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">43.5</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">9.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">18.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">4.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">39.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">21.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">3.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">7.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">13.2</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">BrowseComp Top100</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">39.7</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">13.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">19.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">6.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">33.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">19.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">4.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">6.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">9.7</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">GAIA Text-103</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">88.7</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">49.5</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">47.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">30.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">78.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">57.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">26.5</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">39.5</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">41.1</td></tr>
|
||||
<tr><td colspan="10" style="padding:5px 10px;font-weight:600;color:#1D6FD0;border-bottom:1px solid rgba(29, 111, 208, 0.2);background:rgba(29, 111, 208, 0.14)">通用智能体</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">GDPval-AA v2</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">19.6</strong><sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">4.5</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">11.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.0<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.0</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">Claw-Gym</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>59.2</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">19.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">25.5</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">31.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">51.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">60.0</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">33.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">37.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">2.7</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">WildClaw</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">23.9</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">10.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">9.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">8.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">17.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">20.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">8.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">14.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">4.5</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">QwenClaw</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">42.9</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">19.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">18.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">14.5</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">37.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">36.4</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">16.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">16.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">4.5</td></tr>
|
||||
</tbody></table>
|
||||
<p style="margin:6px 0 0;font-size:11px;line-height:1.55;opacity:0.75">1. <strong style="color:#1D6FD0">蓝色加粗</strong>为该行全场最优结果（含 4B 级模型）；<strong>黑色加粗</strong>为 2B 级模型中的最优结果。<br>2. 带 <sup style="font-size:0.72em;opacity:0.7">†</sup> 的分数取自 Artificial Analysis 官方公布值，其余为内部复现结果。</p>
|
||||
</div>
|
||||
|
||||
## 训练流程
|
||||
|
||||
MiniCPM5-2B 的训练过程是 **[UltraData 分级数据管理体系](https://arxiv.org/pdf/2602.09003)** 的一次完整实践，覆盖 base training、mid-training 与后训练三个阶段。
|
||||
|
||||
**Base training** 采用逐级推进的训练配方，包含 stable training 与 decay training，用于建立基础语言能力与训练稳定性。随后进入 **mid-training**，进一步强化目标能力并适配数据分布。训练语料来自我们同步开源的 [Ultra-FineWeb](https://huggingface.co/datasets/openbmb/Ultra-FineWeb)、[Ultra-FineWeb-L3](https://huggingface.co/datasets/openbmb/Ultra-FineWeb-L3)、[UltraX](https://huggingface.co/datasets/openbmb/UltraX-Preview)、[UltraData-Code](https://huggingface.co/datasets/openbmb/UltraData-Code) 与 [UltraData-Math](https://huggingface.co/datasets/openbmb/UltraData-Math)。
|
||||
|
||||
**后训练阶段**分为 **SFT**、**RL** 与 **OPD** 三步。我们先使用 **400B tokens deep-thinking SFT** 建立深度思考和通用对话能力，相关 SFT 数据已同步开源为 [UltraData-SFT-2605](https://huggingface.co/datasets/openbmb/UltraData-SFT-2605)与[UltraData-SFT-Agent-2609](https://huggingface.co/datasets/openbmb/UltraData-SFT-Agent-2609)。随后针对数学、代码、Agent 和写作等方向训练专用 **RL teacher**（相关数据已同步开源为[UltraData-RL-2609](https://huggingface.co/datasets/openbmb/UltraData-RL-2609)），并通过 **On-Policy Distillation (OPD)** 将这些 teacher 的能力蒸馏回同一个发布模型。
|
||||
|
||||
|
||||
|
||||
### RL + OPD 带来了什么？
|
||||
|
||||
**RL + OPD** 是 MiniCPM5-2B 后训练中的关键环节。**RL** 阶段，使用了 [JustRL II](https://panhaoxuan.notion.site/justrl-ii-small-llms-to-128k-reasoning-with-a-critic-cn) 阐述的 critic-based 算法，大幅提升训练稳定性，并在多个领域取得了显著的收益。在下面列出的基准中，RL + OPD 在推理与通用能力上平均提升 **↑ 10.96 分**，Agent 能力平均提升 **↑ 6.96 分**。
|
||||
|
||||
**OPD** 阶段对 16 个 RL 训练所得到的专家模型（含 5 个 agentic 专家模型）实现了能力合并。训练方式上，我们在 response 序列的每个位置分别对学生模型和教师模型 logits 计算全词表的反向 KL 散度作为优势估计值，替代原有的 verification-based advantage；训练数据上，我们的 OPD 直接复用各 RL teacher 训练时 prompt 作为蒸馏数据，无需额外构造语料。
|
||||
|
||||
|
||||
|
||||
## 快速上手
|
||||
|
||||
> [!Tip]
|
||||
> 我们建议在生成时使用以下采样参数组合：`temperature=1.0, top_p=0.95, min_p=0.0`。
|
||||
>
|
||||
> 如果遇到重复输出，请尝试：`temperature=1.0, top_p=0.95, min_p=0.0, repetition_penalty=1.05`。
|
||||
>
|
||||
> 请注意，不同推理框架对采样参数的支持程度有所差异。
|
||||
|
||||
### vLLM
|
||||
|
||||
```bash
|
||||
pip install "vllm>=0.21"
|
||||
vllm serve openbmb/MiniCPM5-2B --port 8000
|
||||
```
|
||||
|
||||
```bash
|
||||
curl http://localhost:8000/v1/chat/completions \
|
||||
-H "Content-Type: application/json" \
|
||||
-d '{
|
||||
"model": "openbmb/MiniCPM5-2B",
|
||||
"messages": [{"role": "user", "content": "你是谁？可以简单介绍一下自己吗？"}],
|
||||
"max_tokens": 128,
|
||||
"temperature": 1.0
|
||||
}'
|
||||
```
|
||||
|
||||
### SGLang
|
||||
|
||||
```bash
|
||||
pip install "sglang[srt]>=0.5.16"
|
||||
python -m sglang.launch_server --model-path openbmb/MiniCPM5-2B --port 30000
|
||||
```
|
||||
|
||||
```bash
|
||||
curl http://localhost:30000/v1/chat/completions \
|
||||
-H "Content-Type: application/json" \
|
||||
-d '{
|
||||
"model": "openbmb/MiniCPM5-2B",
|
||||
"messages": [{"role": "user", "content": "你是谁？可以简单介绍一下自己吗？"}],
|
||||
"max_tokens": 128,
|
||||
"temperature": 1.0
|
||||
}'
|
||||
```
|
||||
|
||||
**投机采样（DSpark）**：我们同步开源了为 MiniCPM5-2B 训练的 DSpark 草稿模型 [MiniCPM5-2B-DSpark](https://huggingface.co/openbmb/MiniCPM5-2B-DSpark)。在 SGLang 中启用后可加速解码，且不改变目标模型的输出：
|
||||
|
||||
```bash
|
||||
python -m sglang.launch_server \
|
||||
--model-path openbmb/MiniCPM5-2B \
|
||||
--trust-remote-code \
|
||||
--speculative-algorithm DSPARK \
|
||||
--speculative-draft-model-path openbmb/MiniCPM5-2B-DSpark \
|
||||
--speculative-dspark-block-size 7 \
|
||||
--port 30000
|
||||
```
|
||||
|
||||
### Llama.cpp
|
||||
|
||||
```bash
|
||||
llama-server -m MiniCPM5-2B-F16.gguf -a MiniCPM5-2B --port 8080 -ngl 99 -c 8192 --jinja
|
||||
```
|
||||
|
||||
在这个例子里 `-c 8192` 设置了上下文长度为 8192，可以根据需要修改上下文长度。
|
||||
|
||||
```bash
|
||||
curl http://localhost:8080/v1/chat/completions \
|
||||
-H "Content-Type: application/json" \
|
||||
-d '{
|
||||
"model": "MiniCPM5-2B",
|
||||
"messages": [{"role": "user", "content": "1+1=?"}],
|
||||
"temperature": 1.0, "top_p": 0.95, "min_p": 0.0, "max_tokens": 256
|
||||
}'
|
||||
```
|
||||
|
||||
llama.cpp 默认设置 `min_p=0.05`，会过滤掉概率低于最高概率 token 5% 的 token。这种过滤反而可能引发重复——它恰恰过滤掉了模型摆脱重复循环所需的那些 token。我们明确将 `min_p` 设为 `0.0` 以禁用该过滤。
|
||||
|
||||
### Transformers
|
||||
|
||||
```bash
|
||||
pip install -U "transformers>=5.6" accelerate torch
|
||||
```
|
||||
|
||||
```python
|
||||
from transformers import AutoModelForCausalLM, AutoTokenizer
|
||||
model_id = "openbmb/MiniCPM5-2B"
|
||||
tokenizer = AutoTokenizer.from_pretrained(model_id)
|
||||
model = AutoModelForCausalLM.from_pretrained(
|
||||
model_id,
|
||||
torch_dtype="auto",
|
||||
device_map="auto",
|
||||
)
|
||||
messages = [{"role": "user", "content": "你是谁？可以简单介绍一下自己吗？"}]
|
||||
inputs = tokenizer.apply_chat_template(
|
||||
messages,
|
||||
tokenize=True,
|
||||
add_generation_prompt=True,
|
||||
enable_thinking=True,
|
||||
return_dict=True,
|
||||
return_tensors="pt",
|
||||
).to(model.device)
|
||||
outputs = model.generate(**inputs, max_new_tokens=128)
|
||||
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:], skip_special_tokens=True))
|
||||
```
|
||||
|
||||
## 工具调用
|
||||
|
||||
工具调用**推荐使用 SGLang**。MiniCPM5-2B 以 XML 格式产出工具调用，SGLang 内置的 `minicpm5` parser 会自动将其转换为 OpenAI 兼容的 `tool_calls` 字段。
|
||||
|
||||
```bash
|
||||
python -m sglang.launch_server --model-path openbmb/MiniCPM5-2B --port 30000 \
|
||||
--tool-call-parser minicpm5 # 或：--tool-call-parser auto
|
||||
```
|
||||
|
||||
## GitHub Cookbooks 与 Agent Skills
|
||||
|
||||
MiniCPM5-2B 使用**标准** `LlamaForCausalLM` **架构**，主流推理引擎可直接加载，**无需自定义算子，也无模型代码 fork**。逐步部署和微调说明请参考下方 GitHub cookbooks；Agent Skills 作为 GitHub 资源提供给使用 Cursor / Claude Code 类 coding agent 的用户。
|
||||
|
||||
### 部署
|
||||
|
||||
| 后端 | 模型格式 / 适用场景 | Cookbook | Agent Skill |
|
||||
| ------------ | ------------------------------------------------------------- | ----------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------- |
|
||||
| Transformers | BF16 / FP16，本地 Python 推理，GPU + CPU | [transformers.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/deployment/transformers.md) | [minicpm5-deploy-transformers](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-deploy-transformers/SKILL.md) |
|
||||
| vLLM | BF16 / FP16 OpenAI server | [vllm.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/deployment/vllm.md) | [minicpm5-deploy-vllm](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-deploy-vllm/SKILL.md) |
|
||||
| SGLang | BF16 / FP16 OpenAI server，推荐用于 tool calling | [sglang.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/deployment/sglang.md) | [minicpm5-deploy-sglang](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-deploy-sglang/SKILL.md) |
|
||||
| llama.cpp | GGUF，CPU/GPU 本地推理 | [llama_cpp.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/deployment/llama_cpp.md) | [minicpm5-deploy-llama-cpp](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-deploy-llama-cpp/SKILL.md) |
|
||||
| Ollama | GGUF，本地端侧运行 | [ollama.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/deployment/ollama.md) | [minicpm5-deploy-ollama](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-deploy-ollama/SKILL.md) |
|
||||
| LM Studio | GGUF，Mac 桌面应用与 OpenAI server | [lmstudio.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/deployment/lmstudio.md) | [minicpm5-deploy-lmstudio](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-deploy-lmstudio/SKILL.md) |
|
||||
| MLX | MLX / 4bit，Apple Silicon 本地推理 | [mlx.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/deployment/mlx.md) | [minicpm5-deploy-mlx](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-deploy-mlx/SKILL.md) |
|
||||
| ArcLight | GGUF 本地端侧 / CPU / 桌面 / 服务器 | [arclight.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/deployment/arclight.md) | [minicpm5-deploy-arclight](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-deploy-arclight/SKILL.md) |
|
||||
| vLLM Ascend | BF16 / FP16 OpenAI server | [vllm_ascend.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/deployment/vllm_ascend.md) | [minicpm5-deploy-vllm-ascend](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-deploy-vllm-ascend/SKILL.md) |
|
||||
| LiteRT-LM | `.litertlm` 端侧运行时：Android / iOS / 桌面 / IoT，CPU + GPU | [litert.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/deployment/litert.md) | [minicpm5-deploy-litert](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-deploy-litert/SKILL.md) |
|
||||
|
||||
### 微调
|
||||
|
||||
| 框架 | 适用场景 | Cookbook | Agent Skill |
|
||||
| ------------- | --------------- | --------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------- |
|
||||
| TRL + PEFT | LoRA / SFT 微调 | [trl.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/finetune/trl.md) | [minicpm5-finetune-trl](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-finetune-trl/SKILL.md) |
|
||||
| LLaMA-Factory | 微调 | [llamafactory.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/finetune/llamafactory.md) | [minicpm5-finetune-llamafactory](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-finetune-llamafactory/SKILL.md) |
|
||||
| ms-swift | 微调 | [ms_swift.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/finetune/ms_swift.md) | [minicpm5-finetune-ms-swift](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-finetune-ms-swift/SKILL.md) |
|
||||
| unsloth | 微调 | [unsloth.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/finetune/unsloth.md) | [minicpm5-finetune-unsloth](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-finetune-unsloth/SKILL.md) |
|
||||
|
||||
### 其他支持的框架
|
||||
|
||||
除上文列出的部署与微调框架外，MiniCPM5-2B 也支持通过 FlagOS 进行多芯片部署。
|
||||
|
||||
#### FlagOS 介绍
|
||||
|
||||
为解决不同 AI 芯片大规模落地应用，北京智源研究院联合众多科研机构、芯片企业、系统厂商、算法和软件相关单位等国内外机构共同发起并创立了 FlagOS 开源社区。
|
||||
|
||||
FlagOS 社区致力于打造面向多种 AI 芯片的统一、开源的系统软件栈，包括大型算子库、统一AI编译器、并行训推框架、统一通信库等核心开源项目，构建「模型-系统-芯片」三层贯通的开放技术生态，通过“一次开发跨芯迁移”释放硬件计算潜力，打破不同芯片软件栈之间生态隔离，有效降低开发者的迁移成本。FlagOS 社区构建人工智能软硬件生态，突破单一闭源垄断，推动AI硬件技术大范围落地发展，立足中国、拥抱全球合作。
|
||||
|
||||
官网速递：[https://flagos.io](https://flagos.io/)
|
||||
|
||||
<details>
|
||||
<summary>FlagOS 多 AI 芯片支持与使用方式</summary>
|
||||
|
||||
#### FlagOS 多 AI 芯片支持
|
||||
|
||||
基于 FlagOS 极短时间内适配 MiniCPM5-2B 到 9 种不同的 AI 芯片，得益于众智 FlagOS 的多芯片统一 AI 系统软件栈的能力。目前，在 FlagOS 团队构建的面向多架构人工智能芯片的大模型自动迁移、适配与发布平台 FlagRelease 上，已发布 MiniCPM5-2B 的多芯片版本。细节如下：
|
||||
|
||||
| Vendor | ModelScope | Huggingface |
|
||||
| --------- | --------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------- |
|
||||
| Nvidia | [MiniCPM5-2B-nvidia-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-nvidia-FlagOS) | [MiniCPM5-2B-nvidia-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-nvidia-FlagOS) |
|
||||
| Hygon | [MiniCPM5-2B-hygon-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-hygon-FlagOS) | [MiniCPM5-2B-hygon-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-hygon-FlagOS) |
|
||||
| Metax | [MiniCPM5-2B-metax-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-metax-FlagOS) | [MiniCPM5-2B-metax-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-metax-FlagOS) |
|
||||
| Iluvatar | [MiniCPM5-2B-iluvatar-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-iluvatar-FlagOS) | [MiniCPM5-2B-iluvatar-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-iluvatar-FlagOS) |
|
||||
| Zhenwu | [MiniCPM5-2B-zhenwu-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-zhenwu-FlagOS) | [MiniCPM5-2B-zhenwu-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-zhenwu-FlagOS) |
|
||||
| Mthreads | [MiniCPM5-2B-mthreads-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-mthreads-FlagOS) | [MiniCPM5-2B-mthreads-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-mthreads-FlagOS) |
|
||||
| Kunlunxin | [MiniCPM5-2B-kunlunxin-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-kunlunxin-FlagOS) | [MiniCPM5-2B-kunlunxin-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-kunlunxin-FlagOS) |
|
||||
| Ascend | [MiniCPM5-2B-ascend-FlagOS](https://modelscope.cn/models/FlagRelease/MiniCPM5-2B-ascend-FlagOS) | [MiniCPM5-2B-ascend-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-ascend-FlagOS) |
|
||||
| ARM-v9 | [MiniCPM5-2B-Armv9-FlagOS](https://modelscope.cn/models/FlagRelease/MiniCPM5-2B-Armv9-FlagOS) | [MiniCPM5-2B-Armv9-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-Armv9-FlagOS) |
|
||||
|
||||
#### FlagOS 使用方式
|
||||
|
||||
##### 使用 FlagOS 在 Nvidia 体验性能加速
|
||||
|
||||
###### From FlagRelease（**推荐**）
|
||||
|
||||
FlagRelease是FlagOS团队构建的一套面向多架构人工智能芯片的大模型自动迁移、适配与发布平台，已发布MiniCPM5-2B的多芯片版本。FlagRelease 已内置相关软件包，无需用户安装。
|
||||
|
||||
###### FlagRelease 镜像关键版本信息
|
||||
|
||||
###### FlagRelease 使用速递
|
||||
|
||||
| Vendor | ModelScope | Huggingface |
|
||||
| --------- | --------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------- |
|
||||
| Nvidia | [MiniCPM5-2B-nvidia-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-nvidia-FlagOS) | [MiniCPM5-2B-nvidia-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-nvidia-FlagOS) |
|
||||
| Hygon | [MiniCPM5-2B-hygon-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-hygon-FlagOS) | [MiniCPM5-2B-hygon-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-hygon-FlagOS) |
|
||||
| Metax | [MiniCPM5-2B-metax-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-metax-FlagOS) | [MiniCPM5-2B-metax-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-metax-FlagOS) |
|
||||
| Iluvatar | [MiniCPM5-2B-iluvatar-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-iluvatar-FlagOS) | [MiniCPM5-2B-iluvatar-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-iluvatar-FlagOS) |
|
||||
| Zhenwu | [MiniCPM5-2B-zhenwu-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-zhenwu-FlagOS) | [MiniCPM5-2B-zhenwu-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-zhenwu-FlagOS) |
|
||||
| Mthreads | [MiniCPM5-2B-mthreads-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-mthreads-FlagOS) | [MiniCPM5-2B-mthreads-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-mthreads-FlagOS) |
|
||||
| Kunlunxin | [MiniCPM5-2B-kunlunxin-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-kunlunxin-FlagOS) | [MiniCPM5-2B-kunlunxin-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-kunlunxin-FlagOS) |
|
||||
| Ascend | [MiniCPM5-2B-ascend-FlagOS](https://modelscope.cn/models/FlagRelease/MiniCPM5-2B-ascend-FlagOS) | [MiniCPM5-2B-ascend-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-ascend-FlagOS) |
|
||||
| ARM-v9 | [MiniCPM5-2B-Armv9-FlagOS](https://modelscope.cn/models/FlagRelease/MiniCPM5-2B-Armv9-FlagOS) | [MiniCPM5-2B-Armv9-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-Armv9-FlagOS) |
|
||||
|
||||
###### 从零开始
|
||||
|
||||
- 依赖Python3.12, GLIBC_2.39, GLIBCXX_3.4.33, CXXABI_1.3.15 环境
|
||||
|
||||
###### Vllm 版本
|
||||
|
||||
###### 安装 FlagOS 算子库
|
||||
|
||||
官方仓库：[https://github.com/flagos-ai/FlagGems](https://github.com/flagos-ai/FlagGems)
|
||||
|
||||
```PowerShell
|
||||
pip install flag-gems==4.2.1rc0
|
||||
pip install triton==3.5.1
|
||||
```
|
||||
|
||||
###### 开启加速
|
||||
|
||||
通过在vllm执行推理的源码中增加flagGems的导入即可开启flagGems加速
|
||||
|
||||
```Bash
|
||||
import flag_gems
|
||||
flag_gems.enable(record=True, once=True, path="/root/gems.txt")
|
||||
```
|
||||
|
||||
```Bash
|
||||
vllm serve ${model_path} \
|
||||
--trust-remote-code \
|
||||
--dtype bfloat16 \
|
||||
--enforce-eager \
|
||||
--port ${Port} \
|
||||
--served-model-name ${model_name} \
|
||||
--gpu-memory-utilization 0.85
|
||||
```
|
||||
|
||||
##### 使用 FlagOS 统一多芯片后端插件
|
||||
|
||||
**[vllm-plugin-FL](https://github.com/flagos-ai/vllm-plugin-FL)** 是一个为 **vLLM** 推理/服务框架构建的插件，它基于 **FlagOS 的统一多芯片后端**开发，旨在扩展 vLLM 在多种硬件环境下的功能和性能表现。
|
||||
|
||||
###### vllm-plugin-FL 使用
|
||||
|
||||
| 厂商 | 从零开始 | 从 FlagRelease 开始 | |
|
||||
| ------ | -------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------ | ----------------------------------------------------------------------------------------- |
|
||||
| 英伟达 | [vllm-plugin-FL/MiniCPM5-2B](https://github.com/flagos-ai/vllm-plugin-FL/blob/main/examples/minicpm/README.md) | [MiniCPM5-2B-ModelScope](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-nvidia-FlagOS) | [MiniCPM5-2B-nvidia-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-nvidia-FlagOS) |
|
||||
|
||||
</details>
|
||||
|
||||
## 局限性与免责声明
|
||||
|
||||
本模型不具备自主意识或法律主体资格，其输出仅为基于统计模式的文本生成结果，可能不准确、有偏见或具冒犯性，也可能被精心设计的提示词（“越狱”）操纵而产生不符合预期的内容。对政治、健康、金融、法律等敏感话题的回答未经专家审核，不应视为专业建议。
|
||||
|
||||
本模型按“**现状**”提供，不附带任何明示或默示担保，开发者不对使用本模型产生的任何损害承担责任。使用者应仅将模型用于合法、合规且符合伦理的目的，自行配置必要的安全措施，并按当地要求标识 AI 生成内容；不得故意越狱、注入攻击或诱导模型产生有害内容，若进行此类测试，风险自担。
|
||||
|
||||
## 开源协议
|
||||
|
||||
MiniCPM 模型权重与相关代码依照 [Apache-2.0](https://github.com/OpenBMB/MiniCPM/blob/main/LICENSE) 协议发布。
|
||||
|
||||
## 引用
|
||||
|
||||
如果觉得我们的工作有帮助，请引用：
|
||||
|
||||
```bibtex
|
||||
@article{minicpm4,
|
||||
title={Minicpm4: Ultra-efficient llms on end devices},
|
||||
author={MiniCPM, Team},
|
||||
journal={arXiv preprint arXiv:2506.07900},
|
||||
year={2025}
|
||||
}
|
||||
```
|
||||
544
README.md
Normal file
544
README.md
Normal file
@@ -0,0 +1,544 @@
|
||||
---
|
||||
license: apache-2.0
|
||||
language:
|
||||
- en
|
||||
- zh
|
||||
library_name: transformers
|
||||
pipeline_tag: text-generation
|
||||
tags:
|
||||
- minicpm
|
||||
- minicpm5
|
||||
- llama
|
||||
- text-generation
|
||||
- long-context
|
||||
- tool-calling
|
||||
- on-device
|
||||
- edge-ai
|
||||
datasets:
|
||||
- openbmb/Ultra-FineWeb
|
||||
- openbmb/UltraX-Preview
|
||||
- openbmb/Ultra-FineWeb-L3
|
||||
- openbmb/UltraData-Math
|
||||
- openbmb/UltraData-Code
|
||||
- openbmb/UltraData-SFT-2605
|
||||
- openbmb/UltraData-SFT-Agent-2609
|
||||
- openbmb/UltraData-RL-2609
|
||||
---
|
||||
|
||||
<div align="center">
|
||||
<img src="https://raw.githubusercontent.com/OpenBMB/MiniCPM/main/assets/minicpm_logo.png" width="500em" />
|
||||
</div>
|
||||
|
||||
<p align="center">
|
||||
<a href="https://arxiv.org/pdf/2506.07900" target="_blank">MiniCPM Tech Report</a> |
|
||||
<a href="https://modelbest.feishu.cn/wiki/UtWxwcERfiRIpIkBOjuc3h9tn1D" target="_blank">MiniCPM Wiki(Chinese)</a> |
|
||||
<a href="https://github.com/OpenBMB/MiniCPM" target="_blank">GitHub Repo</a> |
|
||||
<a href="https://ultradata.openbmb.cn/" target="_blank">UltraData</a> |
|
||||
<a href="https://huggingface.co/spaces/openbmb/MiniCPM5-2B-Demo" target="_blank">Online Demo</a>
|
||||
</p>
|
||||
|
||||
<p align="center">
|
||||
English |
|
||||
<a href="https://huggingface.co/openbmb/MiniCPM5-2B/blob/main/README-cn.md" target="_blank">中文</a>
|
||||
</p>
|
||||
|
||||
## Highlights
|
||||
|
||||
We are releasing **MiniCPM5-2B**, the second model in the **MiniCPM5** series, following [MiniCPM5-1B](https://huggingface.co/openbmb/MiniCPM5-1B). It is a dense 2B Transformer that scales up the same training recipe, built for on-device, local deployment, and resource-constrained scenarios, reaching 2B-class open-source SOTA.
|
||||
|
||||
🏆 **2B-class open-source SOTA**: compared with strong open-source models of similar size, MiniCPM5-2B achieves SOTA performance within this comparison set. It remains competitive with 4B-class models overall, while showing its advantages over models of comparable size in coding, mathematics, long-context understanding, tool use, and agentic tasks.
|
||||
|
||||
<div id="capability-comparison-radar" class="radar-visual" role="img" aria-label="Capability radar chart comparing MiniCPM5-2B with 4B-class models. Each axis is normalized independently to 100 percent.">
|
||||
<style>
|
||||
#capability-comparison-radar {
|
||||
--foreground: #171717;
|
||||
display: block; width: 100%; max-width: 620px; margin: 0 auto; background: #fff;
|
||||
font-family: -apple-system, BlinkMacSystemFont, "Segoe UI", Roboto, "PingFang SC", "Hiragino Sans GB", "Microsoft YaHei", sans-serif;
|
||||
}
|
||||
#capability-comparison-radar svg { display: block; width: 100%; height: auto; background: #fff; }
|
||||
#capability-comparison-radar .title { fill: var(--foreground); font-size: 16px; font-weight: 600; letter-spacing: 0; }
|
||||
#capability-comparison-radar .axis-label { fill: #333; font-size: 12px; font-weight: 600; }
|
||||
#capability-comparison-radar .ring { fill: none; stroke: rgba(128,128,128,.18); stroke-width: .8; }
|
||||
#capability-comparison-radar .spoke { stroke: rgba(128,128,128,.25); stroke-width: .8; }
|
||||
#capability-comparison-radar .ring-label { fill: #8a8a8a; font-size: 9px; }
|
||||
#capability-comparison-radar .series { stroke-linejoin: round; }
|
||||
#capability-comparison-radar .series.primary { stroke-width: 2.4; }
|
||||
#capability-comparison-radar .legend-label { fill: #171717; font-size: 12px; font-weight: 600; }
|
||||
#capability-comparison-radar .legend-average { fill: #666; font-size: 11px; }
|
||||
#capability-comparison-radar .legend-swatch { rx: 3; }
|
||||
@media (max-width: 900px) { #capability-comparison-radar { overflow-x: auto; } #capability-comparison-radar svg { min-width: 620px; } }
|
||||
</style>
|
||||
<svg viewBox="0 0 680 560" aria-hidden="true">
|
||||
<g transform="translate(-20 0)">
|
||||
<text x="40" y="28" class="title" text-anchor="start">Capability Radar by Dimension</text>
|
||||
<polygon points="340.0,235.0 362.5,243.2 374.5,263.9 370.3,287.5 352.0,302.9 328.0,302.9 309.7,287.5 305.5,263.9 317.5,243.2" class="ring" stroke="rgba(128,128,128,.18)" stroke-width=".8"/>
|
||||
<text x="344.0" y="238.0" class="ring-label">20%</text>
|
||||
<polygon points="340.0,200.0 385.0,216.4 408.9,257.8 400.6,305.0 363.9,335.8 316.1,335.8 279.4,305.0 271.1,257.8 295.0,216.4" class="ring" stroke="rgba(128,128,128,.18)" stroke-width=".8"/>
|
||||
<text x="344.0" y="203.0" class="ring-label">40%</text>
|
||||
<polygon points="340.0,165.0 407.5,189.6 443.4,251.8 430.9,322.5 375.9,368.7 304.1,368.7 249.1,322.5 236.6,251.8 272.5,189.6" class="ring" stroke="rgba(128,128,128,.18)" stroke-width=".8"/>
|
||||
<text x="344.0" y="168.0" class="ring-label">60%</text>
|
||||
<polygon points="340.0,130.0 430.0,162.8 477.9,245.7 461.2,340.0 387.9,401.6 292.1,401.6 218.8,340.0 202.1,245.7 250.0,162.8" class="ring" stroke="rgba(128,128,128,.18)" stroke-width=".8"/>
|
||||
<text x="344.0" y="133.0" class="ring-label">80%</text>
|
||||
<polygon points="340.0,95.0 452.5,135.9 512.3,239.6 491.6,357.5 399.9,434.4 280.1,434.4 188.4,357.5 167.7,239.6 227.5,135.9" class="ring" stroke="rgba(29,111,208,.28)" stroke-width="1"/>
|
||||
<text x="344.0" y="98.0" class="ring-label">100%</text>
|
||||
<line x1="340" y1="270" x2="340.0" y2="95.0" class="spoke"/>
|
||||
<text x="340.0" y="63.5" class="axis-label" text-anchor="middle">Code Reasoning</text>
|
||||
<line x1="340" y1="270" x2="452.5" y2="135.9" class="spoke"/>
|
||||
<text x="472.7" y="111.8" class="axis-label" text-anchor="start">Math Reasoning</text>
|
||||
<line x1="340" y1="270" x2="512.3" y2="239.6" class="spoke"/>
|
||||
<text x="543.4" y="234.1" class="axis-label" text-anchor="start">Instruction Following</text>
|
||||
<line x1="340" y1="270" x2="491.6" y2="357.5" class="spoke"/>
|
||||
<text x="518.8" y="373.2" class="axis-label" text-anchor="start">General Knowledge</text>
|
||||
<line x1="340" y1="270" x2="399.9" y2="434.4" class="spoke"/>
|
||||
<text x="410.6" y="464.0" class="axis-label" text-anchor="start">Long Context</text>
|
||||
<line x1="340" y1="270" x2="280.1" y2="434.4" class="spoke"/>
|
||||
<text x="269.4" y="464.0" class="axis-label" text-anchor="end">Tool Use</text>
|
||||
<line x1="340" y1="270" x2="188.4" y2="357.5" class="spoke"/>
|
||||
<text x="161.2" y="373.3" class="axis-label" text-anchor="end">Coding Agent</text>
|
||||
<line x1="340" y1="270" x2="167.7" y2="239.6" class="spoke"/>
|
||||
<text x="136.6" y="234.1" class="axis-label" text-anchor="end">Search Agent</text>
|
||||
<line x1="340" y1="270" x2="227.5" y2="135.9" class="spoke"/>
|
||||
<text x="207.3" y="111.8" class="axis-label" text-anchor="end">General Agent</text>
|
||||
<polygon points="340.0,136.7 450.1,138.8 498.4,242.1 491.6,357.5 398.4,430.3 289.5,408.8 188.4,357.5 188.2,243.2 249.3,161.9" class="series" fill="#E85D4C" fill-opacity=".10" stroke="#E85D4C" stroke-width="1.5"/>
|
||||
<circle cx="340.0" cy="136.7" r="3" fill="#E85D4C"/>
|
||||
<circle cx="450.1" cy="138.8" r="3" fill="#E85D4C"/>
|
||||
<circle cx="498.4" cy="242.1" r="3" fill="#E85D4C"/>
|
||||
<circle cx="491.6" cy="357.5" r="3" fill="#E85D4C"/>
|
||||
<circle cx="398.4" cy="430.3" r="3" fill="#E85D4C"/>
|
||||
<circle cx="289.5" cy="408.8" r="3" fill="#E85D4C"/>
|
||||
<circle cx="188.4" cy="357.5" r="3" fill="#E85D4C"/>
|
||||
<circle cx="188.2" cy="243.2" r="3" fill="#E85D4C"/>
|
||||
<circle cx="249.3" cy="161.9" r="3" fill="#E85D4C"/>
|
||||
<polygon points="340.0,134.4 448.9,140.2 512.3,239.6 462.2,340.5 367.8,346.5 308.0,358.0 231.0,332.9 242.3,252.8 250.1,162.9" class="series" fill="#2A9D8F" fill-opacity=".10" stroke="#2A9D8F" stroke-width="1.5"/>
|
||||
<circle cx="340.0" cy="134.4" r="3" fill="#2A9D8F"/>
|
||||
<circle cx="448.9" cy="140.2" r="3" fill="#2A9D8F"/>
|
||||
<circle cx="512.3" cy="239.6" r="3" fill="#2A9D8F"/>
|
||||
<circle cx="462.2" cy="340.5" r="3" fill="#2A9D8F"/>
|
||||
<circle cx="367.8" cy="346.5" r="3" fill="#2A9D8F"/>
|
||||
<circle cx="308.0" cy="358.0" r="3" fill="#2A9D8F"/>
|
||||
<circle cx="231.0" cy="332.9" r="3" fill="#2A9D8F"/>
|
||||
<circle cx="242.3" cy="252.8" r="3" fill="#2A9D8F"/>
|
||||
<circle cx="250.1" cy="162.9" r="3" fill="#2A9D8F"/>
|
||||
<polygon points="340.0,189.3 411.4,184.9 502.9,241.3 455.3,336.6 356.7,315.9 288.5,411.4 320.8,281.1 266.8,257.1 298.8,220.9" class="series" fill="#E09F3E" fill-opacity=".10" stroke="#E09F3E" stroke-width="1.5"/>
|
||||
<circle cx="340.0" cy="189.3" r="3" fill="#E09F3E"/>
|
||||
<circle cx="411.4" cy="184.9" r="3" fill="#E09F3E"/>
|
||||
<circle cx="502.9" cy="241.3" r="3" fill="#E09F3E"/>
|
||||
<circle cx="455.3" cy="336.6" r="3" fill="#E09F3E"/>
|
||||
<circle cx="356.7" cy="315.9" r="3" fill="#E09F3E"/>
|
||||
<circle cx="288.5" cy="411.4" r="3" fill="#E09F3E"/>
|
||||
<circle cx="320.8" cy="281.1" r="3" fill="#E09F3E"/>
|
||||
<circle cx="266.8" cy="257.1" r="3" fill="#E09F3E"/>
|
||||
<circle cx="298.8" cy="220.9" r="3" fill="#E09F3E"/>
|
||||
<polygon points="340.0,95.0 452.5,135.9 499.8,241.8 476.2,348.7 399.9,434.4 280.1,434.4 220.0,339.3 167.7,239.6 227.5,135.9" class="series primary" fill="#1D6FD0" fill-opacity=".22" stroke="#1D6FD0" stroke-width="2.4"/>
|
||||
<circle cx="340.0" cy="95.0" r="3" fill="#1D6FD0"/>
|
||||
<circle cx="452.5" cy="135.9" r="3" fill="#1D6FD0"/>
|
||||
<circle cx="499.8" cy="241.8" r="3" fill="#1D6FD0"/>
|
||||
<circle cx="476.2" cy="348.7" r="3" fill="#1D6FD0"/>
|
||||
<circle cx="399.9" cy="434.4" r="3" fill="#1D6FD0"/>
|
||||
<circle cx="280.1" cy="434.4" r="3" fill="#1D6FD0"/>
|
||||
<circle cx="220.0" cy="339.3" r="3" fill="#1D6FD0"/>
|
||||
<circle cx="167.7" cy="239.6" r="3" fill="#1D6FD0"/>
|
||||
<circle cx="227.5" cy="135.9" r="3" fill="#1D6FD0"/>
|
||||
<rect x="40" y="490" width="18" height="18" class="legend-swatch" fill="#1D6FD0"/>
|
||||
<text x="66" y="504" class="legend-label">MiniCPM5-2B</text>
|
||||
<text x="66" y="520" class="legend-average">avg 53.9</text>
|
||||
<rect x="195" y="490" width="18" height="18" class="legend-swatch" fill="#E85D4C"/>
|
||||
<text x="221" y="504" class="legend-label">Qwen3.5-4B</text>
|
||||
<text x="221" y="520" class="legend-average">avg 51.1</text>
|
||||
<rect x="350" y="490" width="18" height="18" class="legend-swatch" fill="#2A9D8F"/>
|
||||
<text x="376" y="504" class="legend-label">granite-4.2-3B</text>
|
||||
<text x="376" y="520" class="legend-average">avg 42.7</text>
|
||||
<rect x="505" y="490" width="18" height="18" class="legend-swatch" fill="#E09F3E"/>
|
||||
<text x="531" y="504" class="legend-label">LFM2.5-2.6B</text>
|
||||
<text x="531" y="520" class="legend-average">avg 33.2</text>
|
||||
<text x="340" y="548" class="legend-average" text-anchor="middle">each axis: max = 100%</text>
|
||||
</g>
|
||||
</svg>
|
||||
</div>
|
||||
|
||||
📂 **Open High-Quality Data**: Alongside the model, we are releasing the high-quality training datasets behind it as part of the [UltraData](https://ultradata.openbmb.cn/) family: [UltraX](https://huggingface.co/datasets/openbmb/UltraX-Preview), a high-quality web pre-training dataset; [UltraData-Code](https://huggingface.co/datasets/openbmb/UltraData-Code), featuring L0–L3 tiered code data management to drive a significant leap in coding capabilities; [UltraData-SFT-Agent-2609](https://huggingface.co/datasets/openbmb/UltraData-SFT-Agent-2609), comprising 500K agent training samples to enhance comprehensive on-device agent capabilities; and [UltraData-RL-2609](https://huggingface.co/datasets/openbmb/UltraData-RL-2609), with 80K+ high-quality RL training samples covering mathematics, code, general knowledge, and long-context reasoning.
|
||||
|
||||
## Model List
|
||||
|
||||
Use this directory to choose the model format that matches your runtime:
|
||||
|
||||
**MiniCPM5-2B**
|
||||
|
||||
- **[MiniCPM5-2B](https://huggingface.co/openbmb/MiniCPM5-2B)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-2B) · BF16 final release (post-trained with RL + OPD)
|
||||
- **[MiniCPM5-2B-SFT](https://huggingface.co/openbmb/MiniCPM5-2B-SFT)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-2B-SFT) · BF16 SFT-only checkpoint (before RL / OPD)
|
||||
- **[MiniCPM5-2B-Midtrain](https://huggingface.co/openbmb/MiniCPM5-2B-Midtrain)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-2B-Midtrain) · BF16 mid-training checkpoint (before SFT)
|
||||
- **[MiniCPM5-2B-Base](https://huggingface.co/openbmb/MiniCPM5-2B-Base)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-2B-Base) · BF16 base checkpoint (pre-training only) **👈 you are here**
|
||||
- **[MiniCPM5-2B-GGUF](https://huggingface.co/openbmb/MiniCPM5-2B-GGUF)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-2B-GGUF) · GGUF for llama.cpp / Ollama / LM Studio
|
||||
- **[MiniCPM5-2B-MLX](https://huggingface.co/openbmb/MiniCPM5-2B-MLX)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-2B-MLX) · MLX / 4bit for Apple Silicon
|
||||
- **[MiniCPM5-2B-GPTQ](https://huggingface.co/openbmb/MiniCPM5-2B-GPTQ)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-2B-GPTQ) · GPTQ / 4bit quantized model
|
||||
- **[MiniCPM5-2B-DSpark](https://huggingface.co/openbmb/MiniCPM5-2B-DSpark)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-2B-DSpark) · DSpark draft model for inference acceleration
|
||||
- **[MiniCPM5-2B-DSpark-GGUF](https://huggingface.co/openbmb/MiniCPM5-2B-DSpark-GGUF)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-2B-DSpark-GGUF) · GGUF version of DSpark draft model
|
||||
- **[MiniCPM5-2B-LiteRT](https://huggingface.co/litert-community/MiniCPM5-2B)** · [ModelScope](https://www.modelscope.cn/models/litert-community/MiniCPM5-2B) · the LiteRT-LM version of MiniCPM5-2B
|
||||
|
||||
**MiniCPM5-1B**
|
||||
|
||||
- **[MiniCPM5-1B](https://huggingface.co/openbmb/MiniCPM5-1B)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-1B) · BF16 final release (post-trained with RL + OPD)
|
||||
- **[MiniCPM5-1B-SFT](https://huggingface.co/openbmb/MiniCPM5-1B-SFT)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-1B-SFT) · BF16 SFT-only checkpoint (before RL / OPD)
|
||||
- **[MiniCPM5-1B-Base](https://huggingface.co/openbmb/MiniCPM5-1B-Base)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-1B-Base) · BF16 base checkpoint (pre-training only)
|
||||
- **[MiniCPM5-1B-GGUF](https://huggingface.co/openbmb/MiniCPM5-1B-GGUF)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-1B-GGUF) · GGUF for llama.cpp / Ollama / LM Studio
|
||||
- **[MiniCPM5-1B-MLX](https://huggingface.co/openbmb/MiniCPM5-1B-MLX)** · [ModelScope](https://www.modelscope.cn/models/OpenBMB/MiniCPM5-1B-MLX) · MLX / 4bit for Apple Silicon
|
||||
|
||||
## Model Information
|
||||
|
||||
MiniCPM5-2B has the following features:
|
||||
|
||||
- **Type**: Causal Language Model
|
||||
- **Architecture**: Standard `LlamaForCausalLM`
|
||||
- **Number of Parameters**: 2,516,756,480
|
||||
- **Number of Non-Embedding Parameters**: 1,981,982,720
|
||||
- **Number of Layers**: 42
|
||||
- **Number of Attention Heads (GQA)**: 16 for Q and 2 for KV
|
||||
- **Context Length**: 131,072
|
||||
|
||||
## Introduction
|
||||
|
||||
MiniCPM5-2B is the second model in the MiniCPM5 series. It is designed for local assistants, coding agents, tool-use workflows, and reasoning scenarios where a compact model is preferred. The model keeps a small deployment footprint while providing native long-context support.
|
||||
|
||||
## Evaluation Results
|
||||
|
||||
We compare **MiniCPM5-2B** with strong open-source models in the same size class, including **LFM2.5-2.6B**, **Qwen3.5-2B**, and **Gemma-4-E2B-it**, while also listing larger models such as **Qwen3.5-4B**, **granite-4.2-3B**, **Nemotron-3-Nano-4B**, **Gemma-4-E4B-it**, and **LFM2.5-8B-A1B** for reference.
|
||||
|
||||
Within this comparison set, MiniCPM5-2B reaches 2B-class open-source SOTA with an average score of **53.9**, and also exceeds all of the larger models included here (the highest is **51.1**). Its advantages are most visible in code reasoning, math reasoning, long-context understanding, tool use, and multiple agentic tasks.
|
||||
|
||||
<div style="width:100%;max-width:1080px;margin:0 auto;padding:16px 0;background:#fff;
|
||||
font-family:-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,'PingFang SC',
|
||||
'Hiragino Sans GB','Microsoft YaHei',sans-serif;color:#171717">
|
||||
<h1 style="margin:0 0 14px;font-size:22px;font-weight:700;color:#1D6FD0;
|
||||
letter-spacing:0.02em">Evaluation Results of MiniCPM5-2B and Baselines</h1>
|
||||
<table class="vl-table" style="width:100%;margin:0;table-layout:fixed;border-collapse:collapse;font-size:13px;font-variant-numeric:tabular-nums"><thead><tr><th rowspan="2" style="padding:7px 5px;text-align:left;font-weight:600;border-bottom:2px solid #1D6FD0;color:#1D6FD0;width:18%"></th><th rowspan="2" style="padding:7px 4px;text-align:center;font-weight:600;border-bottom:2px solid #1D6FD0;color:#1D6FD0;font-size:12px;width:9.111%;background:rgba(29, 111, 208, 0.08);vertical-align:middle;word-break:normal;">MiniCPM5-2B</th><th colspan="3" style="padding:6px 4px;text-align:center;font-weight:600;color:#1D6FD0;font-size:13px;border-bottom:1px solid rgba(29, 111, 208, 0.2);border-left:1px solid rgba(29, 111, 208, 0.25);">2B-class Models</th><th colspan="5" style="padding:6px 4px;text-align:center;font-weight:600;color:#1D6FD0;font-size:13px;border-bottom:1px solid rgba(29, 111, 208, 0.2);border-left:1px solid rgba(29, 111, 208, 0.25);">4B-class Models</th></tr><tr><th style="padding:6px 4px;text-align:center;font-weight:500;border-bottom:2px solid #1D6FD0;color:#1D6FD0;font-size:11px;width:9.111%;border-left:1px solid rgba(29, 111, 208, 0.25);word-break:normal;vertical-align:middle;">LFM2.5-2.6B</th><th style="padding:6px 4px;text-align:center;font-weight:500;border-bottom:2px solid #1D6FD0;color:#1D6FD0;font-size:11px;width:9.111%;word-break:normal;vertical-align:middle;">Qwen3.5-2B</th><th style="padding:6px 4px;text-align:center;font-weight:500;border-bottom:2px solid #1D6FD0;color:#1D6FD0;font-size:11px;width:9.111%;word-break:normal;vertical-align:middle;">Gemma-4-E2B-it</th><th style="padding:6px 4px;text-align:center;font-weight:500;border-bottom:2px solid #1D6FD0;color:#1D6FD0;font-size:11px;width:9.111%;border-left:1px solid rgba(29, 111, 208, 0.25);word-break:normal;vertical-align:middle;">Qwen3.5-4B</th><th style="padding:6px 4px;text-align:center;font-weight:500;border-bottom:2px solid #1D6FD0;color:#1D6FD0;font-size:11px;width:9.111%;word-break:normal;vertical-align:middle;">granite-4.2-3B</th><th style="padding:6px 4px;text-align:center;font-weight:500;border-bottom:2px solid #1D6FD0;color:#1D6FD0;font-size:11px;width:9.111%;word-break:normal;vertical-align:middle;">Nemotron-3-Nano-4B</th><th style="padding:6px 4px;text-align:center;font-weight:500;border-bottom:2px solid #1D6FD0;color:#1D6FD0;font-size:11px;width:9.111%;word-break:normal;vertical-align:middle;">Gemma-4-E4B-it</th><th style="padding:6px 4px;text-align:center;font-weight:500;border-bottom:2px solid #1D6FD0;color:#1D6FD0;font-size:11px;width:9.111%;word-break:normal;vertical-align:middle;">LFM2.5-8B-A1B</th></tr></thead><tbody>
|
||||
<tr style="background:rgba(29, 111, 208, 0.03)"><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">Average</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;font-weight:600;"><strong style="color:#1D6FD0">53.9</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;font-weight:600;">33.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;font-weight:600;">28.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;font-weight:600;">24.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;font-weight:600;">51.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;font-weight:600;">42.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;font-weight:600;">32.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;font-weight:600;">31.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;font-weight:600;">28.4</td></tr>
|
||||
<tr><td colspan="10" style="padding:5px 10px;font-weight:600;color:#1D6FD0;border-bottom:1px solid rgba(29, 111, 208, 0.2);background:rgba(29, 111, 208, 0.14)">Code Reasoning</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">LiveCodeBench v6</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">69.1</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">42.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">20.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">42.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">56.4</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">58.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">50.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">53.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">39.8</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">LCB-Pro 25Q2 (Easy)</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">68.0</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">30.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">10.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">27.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">58.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">54.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">51.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">45.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">27.8</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">LCB-Pro 25Q2 (Medium)</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">17.5</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">0.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">7.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">5.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">5.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">1.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.0</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">OJBench</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">32.5</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">11.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">2.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">11.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">24.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">21.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">20.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">19.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">8.2</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">SciCode (wbg)</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">26.3</strong><sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">14.2<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">2.8<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">20.9<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">16.1<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">24.9<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">16.4<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">24.4<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">7.8<sup style="font-size:0.72em;opacity:0.7">†</sup></td></tr>
|
||||
<tr><td colspan="10" style="padding:5px 10px;font-weight:600;color:#1D6FD0;border-bottom:1px solid rgba(29, 111, 208, 0.2);background:rgba(29, 111, 208, 0.14)">Math Reasoning</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">AIME 2025</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">86.5</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">41.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">29.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">31.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">78.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">79.4</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">56.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">37.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">46.0</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">AIME 2026</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">86.5</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">45.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">29.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">39.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">82.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">83.5</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">62.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">45.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">56.7</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">HMMT Feb 2026</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>63.8</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">33.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">20.5</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">17.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">64.0</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">60.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">51.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">30.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">38.5</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">MATH-500</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>94.6</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">89.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">85.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">85.4</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">99.0</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">97.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">91.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">88.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">93.2</td></tr>
|
||||
<tr><td colspan="10" style="padding:5px 10px;font-weight:600;color:#1D6FD0;border-bottom:1px solid rgba(29, 111, 208, 0.2);background:rgba(29, 111, 208, 0.14)">Instruction Following</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">IFBench</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>66.3</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">59.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">46.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">25.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">59.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">73.0</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">58.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">28.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">51.0</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">IFEval</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;">86.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>93.4</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">77.5</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">31.4</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">90.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">93.7</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">88.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">44.4</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">90.8</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">Multi-IF</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;">71.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">76.8</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">57.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">40.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">73.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">75.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">65.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">45.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">71.4</td></tr>
|
||||
<tr><td colspan="10" style="padding:5px 10px;font-weight:600;color:#1D6FD0;border-bottom:1px solid rgba(29, 111, 208, 0.2);background:rgba(29, 111, 208, 0.14)">General Knowledge</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">MMLU-Pro</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>70.8</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">65.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">64.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">56.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">78.0</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">65.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">65.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">68.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">63.1</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">MMLU-Redux</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>84.7</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">80.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">80.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">71.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">88.7</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">78.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">79.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">83.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">80.0</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">HLE</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>8.9</strong><sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">6.2<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">2.6<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">4.8<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">9.9</strong><sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">6.6<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">4.9<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">3.8<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">6.9<sup style="font-size:0.72em;opacity:0.7">†</sup></td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">GPQA-Diamond</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>70.2</strong><sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">55.8<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">45.6<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">43.3<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">77.1</strong><sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">55.9<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">51.3<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">57.6<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">51.3<sup style="font-size:0.72em;opacity:0.7">†</sup></td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">SuperGPQA</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>40.8</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">26.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">38.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">30.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">52.8</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">39.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">37.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">38.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">34.5</td></tr>
|
||||
<tr><td colspan="10" style="padding:5px 10px;font-weight:600;color:#1D6FD0;border-bottom:1px solid rgba(29, 111, 208, 0.2);background:rgba(29, 111, 208, 0.14)">Long Context</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">AA-LCR</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>59.0</strong><sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">5.3<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">28.7<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">17.0<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">61.0</strong><sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">24.3<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">17.3<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">33.0<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.0<sup style="font-size:0.72em;opacity:0.7">†</sup></td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">NoLiMa</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">68.1</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">0.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">17.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">3.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">43.5</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">5.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">1.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">2.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.5</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">LongBenchPro</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>44.8</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">23.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">8.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">42.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">58.4</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">34.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">27.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">53.5</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">19.6</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">LongBench v2</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>43.7</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">30.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">24.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">33.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">47.3</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">36.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">32.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">42.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">30.4</td></tr>
|
||||
<tr><td colspan="10" style="padding:5px 10px;font-weight:600;color:#1D6FD0;border-bottom:1px solid rgba(29, 111, 208, 0.2);background:rgba(29, 111, 208, 0.14)">Tool Use</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">τ³-Bench Banking</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">20.8</strong><sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">7.2<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">2.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">3.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">6.8<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">5.6<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">1.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">4.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">3.4</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">τ²-Bench Telecom</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">97.1</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">90.4</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">69.0<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">20.8<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">92.1<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">40.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">28.1<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">20.8<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">16.1<sup style="font-size:0.72em;opacity:0.7">†</sup></td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">BFCL v4</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">66.6</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">61.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">43.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">36.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">56.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">52.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">43.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">47.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">49.2</td></tr>
|
||||
<tr><td colspan="10" style="padding:5px 10px;font-weight:600;color:#1D6FD0;border-bottom:1px solid rgba(29, 111, 208, 0.2);background:rgba(29, 111, 208, 0.14)">Coding Agent</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">SWE-bench Verified</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">46.4</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">6.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">5.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">2.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">33.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">36.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">3.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">15.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.4</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">SWE-bench Pro</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>14.4</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">0.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">28.2</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">12.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">3.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.4</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">Terminal-Bench v2.1</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>8.6</strong><sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">4.5<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">3.0<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.4<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">25.8</strong><sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">13.9<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">3.8<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">1.9<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">1.9</td></tr>
|
||||
<tr><td colspan="10" style="padding:5px 10px;font-weight:600;color:#1D6FD0;border-bottom:1px solid rgba(29, 111, 208, 0.2);background:rgba(29, 111, 208, 0.14)">Search Agent</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">BrowseComp-ZH</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">43.5</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">9.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">18.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">4.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">39.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">21.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">3.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">7.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">13.2</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">BrowseComp Top100</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">39.7</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">13.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">19.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">6.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">33.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">19.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">4.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">6.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">9.7</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">GAIA Text-103</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">88.7</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">49.5</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">47.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">30.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">78.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">57.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">26.5</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">39.5</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">41.1</td></tr>
|
||||
<tr><td colspan="10" style="padding:5px 10px;font-weight:600;color:#1D6FD0;border-bottom:1px solid rgba(29, 111, 208, 0.2);background:rgba(29, 111, 208, 0.14)">General Agent</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">GDPval-AA v2</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">19.6</strong><sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">4.5</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">11.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.0<sup style="font-size:0.72em;opacity:0.7">†</sup></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">0.0</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">Claw-Gym</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong>59.2</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">19.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">25.5</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">31.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">51.6</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">60.0</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">33.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">37.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">2.7</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">WildClaw</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">23.9</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">10.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">9.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">8.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">17.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">20.0</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">8.9</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">14.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">4.5</td></tr>
|
||||
<tr><td class="benchmark-cell" style="padding:5px 5px;padding-left:14px;border-bottom:1px solid rgba(128, 128, 128, 0.15);"><div class="benchmark-capability" style="font-size:14px;font-weight:600;line-height:1.22;color:#171717">QwenClaw</div></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);background:rgba(29, 111, 208, 0.08);vertical-align:middle;font-size:14px;line-height:1.2;"><strong style="color:#1D6FD0">42.9</strong></td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">19.3</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">18.2</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">14.5</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);border-left:1px solid rgba(29, 111, 208, 0.25);vertical-align:middle;font-size:14px;line-height:1.2;">37.1</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">36.4</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">16.8</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">16.7</td><td style="padding:5px 3px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15);vertical-align:middle;font-size:14px;line-height:1.2;">4.5</td></tr>
|
||||
</tbody></table>
|
||||
<p style="margin:6px 0 0;font-size:11px;line-height:1.55;opacity:0.75">1. <strong style="color:#1D6FD0">Blue bold</strong> indicates the best result across all models in the row (including 4B-class models); <strong>Black bold</strong> indicates the best result among 2B-class models.<br>2. Scores marked <sup style="font-size:0.72em;opacity:0.7">†</sup> come from the official Artificial Analysis release; all others are reproduced internally.</p>
|
||||
</div>
|
||||
|
||||
## Training Recipe
|
||||
|
||||
The training of MiniCPM5-2B is a full-stack practice of **[UltraData Tiered Data Management](https://arxiv.org/pdf/2602.09003)**, covering three stages: base training, mid-training, and post-training.
|
||||
|
||||
During **base training**, the model goes through stable training and decay training to build core language capability and training stability. It then enters **mid-training** to further strengthen target capabilities and adapt to the target data distribution. The training corpus is released alongside the model as [Ultra-FineWeb](https://huggingface.co/datasets/openbmb/Ultra-FineWeb), [Ultra-FineWeb-L3](https://huggingface.co/datasets/openbmb/Ultra-FineWeb-L3), [UltraX](https://huggingface.co/datasets/openbmb/UltraX-Preview), [UltraData-Code](https://huggingface.co/datasets/openbmb/UltraData-Code) and [UltraData-Math](https://huggingface.co/datasets/openbmb/UltraData-Math).
|
||||
|
||||
During **post-training**, we proceed in three steps: **SFT**, **RL**, and **OPD**. We first use **400B tokens of deep-thinking SFT** to establish deep-thinking and general chat abilities; the SFT data is released as [UltraData-SFT-2605](https://huggingface.co/datasets/openbmb/UltraData-SFT-2605) and the Agent SFT data is released as [UltraData-SFT-Agent-2609](https://huggingface.co/datasets/openbmb/UltraData-SFT-Agent-2609). We then train specialized **RL teachers** for math, code, agentic tasks, writing, and related domains (with the corresponding data also open-sourced as [UltraData-RL-2609](https://huggingface.co/datasets/openbmb/UltraData-RL-2609)), and use **On-Policy Distillation (OPD)** to distill these teachers back into one release model.
|
||||
|
||||
|
||||
|
||||
### What does RL + OPD bring?
|
||||
|
||||
**RL + OPD** is a key part of MiniCPM5-2B post-training. During the **RL** stage, we adopted the critic-based algorithm described in [JustRL II](https://panhaoxuan.notion.site/justrl-ii-scaling-small-llms-to-128k-reasoning-with-a-critic), substantially improving training stability and achieving significant gains across multiple domains. On the benchmarks listed below, RL + OPD improves reasoning and general capabilities by an average of **↑10.96 points**, and agentic capabilities by **↑6.96 points**.
|
||||
|
||||
**OPD** merges the capabilities of 16 expert models produced by RL training, including 5 agentic expert models. At each response position, we compute the full-vocabulary reverse KL divergence between student and teacher logits as the advantage estimate, replacing the original verification-based advantage. OPD directly reuses the prompts used to train each RL teacher as distillation data, so no additional corpus construction is required.
|
||||
|
||||
|
||||
|
||||
## Quickstart
|
||||
|
||||
> [!Tip]
|
||||
> We recommend using the following sets of sampling parameters for generation: `temperature=1.0, top_p=0.95, min_p=0.0`.
|
||||
>
|
||||
> If you encounter repetitive outputs, try: `temperature=1.0, top_p=0.95, min_p=0.0, repetition_penalty=1.05`.
|
||||
>
|
||||
> Please note that the support for sampling parameters varies according to inference frameworks.
|
||||
|
||||
### vLLM
|
||||
|
||||
```bash
|
||||
pip install "vllm>=0.21"
|
||||
vllm serve openbmb/MiniCPM5-2B --port 8000
|
||||
```
|
||||
|
||||
```bash
|
||||
curl http://localhost:8000/v1/chat/completions \
|
||||
-H "Content-Type: application/json" \
|
||||
-d '{
|
||||
"model": "openbmb/MiniCPM5-2B",
|
||||
"messages": [{"role": "user", "content": "Who are you? Please briefly introduce yourself."}],
|
||||
"max_tokens": 128,
|
||||
"temperature": 1.0
|
||||
}'
|
||||
```
|
||||
|
||||
### SGLang
|
||||
|
||||
```bash
|
||||
pip install "sglang[srt]>=0.5.16"
|
||||
python -m sglang.launch_server --model-path openbmb/MiniCPM5-2B --port 30000
|
||||
```
|
||||
|
||||
```bash
|
||||
curl http://localhost:30000/v1/chat/completions \
|
||||
-H "Content-Type: application/json" \
|
||||
-d '{
|
||||
"model": "openbmb/MiniCPM5-2B",
|
||||
"messages": [{"role": "user", "content": "Who are you? Please briefly introduce yourself."}],
|
||||
"max_tokens": 128,
|
||||
"temperature": 1.0
|
||||
}'
|
||||
```
|
||||
|
||||
**Speculative decoding (DSpark)**: we also release [MiniCPM5-2B-DSpark](https://huggingface.co/openbmb/MiniCPM5-2B-DSpark), a DSpark draft model trained for MiniCPM5-2B. Enable it in SGLang to accelerate decoding while keeping the target model's outputs unchanged:
|
||||
|
||||
```bash
|
||||
python -m sglang.launch_server \
|
||||
--model-path openbmb/MiniCPM5-2B \
|
||||
--trust-remote-code \
|
||||
--speculative-algorithm DSPARK \
|
||||
--speculative-draft-model-path openbmb/MiniCPM5-2B-DSpark \
|
||||
--speculative-dspark-block-size 7 \
|
||||
--port 30000
|
||||
```
|
||||
|
||||
### Llama.cpp
|
||||
|
||||
```bash
|
||||
llama-server -m MiniCPM5-2B-F16.gguf -a MiniCPM5-2B --port 8080 -ngl 99 -c 8192 --jinja
|
||||
```
|
||||
|
||||
`-c 8192` sets the context length. You can adjust this value as needed.
|
||||
|
||||
```bash
|
||||
curl http://localhost:8080/v1/chat/completions \
|
||||
-H "Content-Type: application/json" \
|
||||
-d '{
|
||||
"model": "MiniCPM5-2B",
|
||||
"messages": [{"role": "user", "content": "1+1=?"}],
|
||||
"temperature": 1.0, "top_p": 0.95, "min_p": 0.0, "max_tokens": 256
|
||||
}'
|
||||
```
|
||||
|
||||
In llama.cpp, the default `min_p=0.05` can lead to repetitive output: it filters out tokens whose probability is below 5% of the highest-probability token, potentially discarding the exact tokens needed to break out of a repetition loop. To prevent this, we set `min_p=0.0`.
|
||||
|
||||
### Transformers
|
||||
|
||||
```bash
|
||||
pip install -U "transformers>=5.6" accelerate torch
|
||||
```
|
||||
|
||||
```python
|
||||
from transformers import AutoModelForCausalLM, AutoTokenizer
|
||||
model_id = "openbmb/MiniCPM5-2B"
|
||||
tokenizer = AutoTokenizer.from_pretrained(model_id)
|
||||
model = AutoModelForCausalLM.from_pretrained(
|
||||
model_id,
|
||||
torch_dtype="auto",
|
||||
device_map="auto",
|
||||
)
|
||||
messages = [{"role": "user", "content": "Who are you? Please briefly introduce yourself."}]
|
||||
inputs = tokenizer.apply_chat_template(
|
||||
messages,
|
||||
tokenize=True,
|
||||
add_generation_prompt=True,
|
||||
enable_thinking=True,
|
||||
return_dict=True,
|
||||
return_tensors="pt",
|
||||
).to(model.device)
|
||||
outputs = model.generate(**inputs, max_new_tokens=128)
|
||||
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:], skip_special_tokens=True))
|
||||
```
|
||||
|
||||
## Tool Calling
|
||||
|
||||
For tool / function calling, **SGLang is the recommended backend**. MiniCPM5-2B emits XML-style tool calls and SGLang's built-in `minicpm5` parser converts them to OpenAI-compatible `tool_calls` natively:
|
||||
|
||||
```bash
|
||||
python -m sglang.launch_server --model-path openbmb/MiniCPM5-2B --port 30000 \
|
||||
--tool-call-parser minicpm5 # or: --tool-call-parser auto
|
||||
```
|
||||
|
||||
## GitHub Cookbooks and Agent Skills
|
||||
|
||||
MiniCPM5-2B uses the **standard `LlamaForCausalLM` architecture**, so mainstream inference engines can load it directly: **no custom kernels, no model-code fork**. For step-by-step deployment and fine-tuning instructions, use the GitHub cookbooks below. Agent Skills are linked as GitHub resources for users working with Cursor / Claude Code style coding agents.
|
||||
|
||||
### Deployment
|
||||
|
||||
| Backend | Model format / use case | Cookbook | Agent Skill |
|
||||
| ------------ | ----------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------- |
|
||||
| Transformers | BF16 / FP16 local Python inference, GPU + CPU | [transformers.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/deployment/transformers.md) | [minicpm5-deploy-transformers](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-deploy-transformers/SKILL.md) |
|
||||
| vLLM | BF16 / FP16 OpenAI server | [vllm.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/deployment/vllm.md) | [minicpm5-deploy-vllm](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-deploy-vllm/SKILL.md) |
|
||||
| SGLang | BF16 / FP16 OpenAI server, recommended for tool calling | [sglang.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/deployment/sglang.md) | [minicpm5-deploy-sglang](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-deploy-sglang/SKILL.md) |
|
||||
| llama.cpp | GGUF local inference, CPU/GPU | [llama_cpp.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/deployment/llama_cpp.md) | [minicpm5-deploy-llama-cpp](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-deploy-llama-cpp/SKILL.md) |
|
||||
| Ollama | GGUF local on-device runtime | [ollama.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/deployment/ollama.md) | [minicpm5-deploy-ollama](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-deploy-ollama/SKILL.md) |
|
||||
| LM Studio | GGUF Mac desktop app and OpenAI server | [lmstudio.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/deployment/lmstudio.md) | [minicpm5-deploy-lmstudio](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-deploy-lmstudio/SKILL.md) |
|
||||
| MLX | MLX / 4bit local inference on Apple Silicon | [mlx.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/deployment/mlx.md) | [minicpm5-deploy-mlx](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-deploy-mlx/SKILL.md) |
|
||||
| ArcLight | GGUF local on-device, CPU, Desktop & Server | [arclight.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/deployment/arclight.md) | [minicpm5-deploy-arclight](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-deploy-arclight/SKILL.md) |
|
||||
| vLLM Ascend | BF16 / FP16 OpenAI server | [vllm_ascend.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/deployment/vllm_ascend.md) | [minicpm5-deploy-vllm-ascend](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-deploy-vllm-ascend/SKILL.md) |
|
||||
| LiteRT-LM | `.litertlm` on-device runtime: Android / iOS / desktop / IoT, CPU + GPU | [litert.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/deployment/litert.md) | [minicpm5-deploy-litert](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-deploy-litert/SKILL.md) |
|
||||
|
||||
### Fine-tuning
|
||||
|
||||
| Framework | Use case | Cookbook | Agent Skill |
|
||||
| ------------- | ---------------------- | --------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------- |
|
||||
| TRL + PEFT | LoRA / SFT fine-tuning | [trl.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/finetune/trl.md) | [minicpm5-finetune-trl](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-finetune-trl/SKILL.md) |
|
||||
| LLaMA-Factory | Fine-tuning | [llamafactory.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/finetune/llamafactory.md) | [minicpm5-finetune-llamafactory](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-finetune-llamafactory/SKILL.md) |
|
||||
| ms-swift | Fine-tuning | [ms_swift.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/finetune/ms_swift.md) | [minicpm5-finetune-ms-swift](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-finetune-ms-swift/SKILL.md) |
|
||||
| unsloth | Fine-tuning | [unsloth.md](https://github.com/OpenBMB/MiniCPM/blob/main/docs/finetune/unsloth.md) | [minicpm5-finetune-unsloth](https://github.com/OpenBMB/MiniCPM/blob/main/skills/minicpm5-finetune-unsloth/SKILL.md) |
|
||||
|
||||
### Other Supported Frameworks
|
||||
|
||||
In addition to the deployment and fine-tuning frameworks listed above, MiniCPM5-2B is also supported by FlagOS for multi-chip deployment.
|
||||
|
||||
#### FlagOS Overview
|
||||
|
||||
To enable large-scale deployment across different AI chips, Beijing Zhiyuan Research Institute, together with numerous research institutions, chip manufacturers, system vendors, and algorithm and software organizations both domestically and internationally, jointly initiated and established the FlagOS Open Source Community.
|
||||
|
||||
The FlagOS community is dedicated to building a unified, open-source system software stack for various AI chips, encompassing core open-source projects such as a large-scale operator library, a unified AI compiler, parallel training and inference frameworks, and a unified communication library. It aims to create an open technology ecosystem connecting the “model-system-chip” layers. By enabling “develop once, deploy across chips”, FlagOS unlocks the computational potential of hardware, breaks down the ecosystem silos between different chip software stacks, and effectively reduces migration costs for developers.The FlagOS community fosters an AI hardware and software ecosystem, overcomes single-vendor closed-source monopolies, promotes widespread deployment of AI hardware technologies, and is committed to rooted in China while embracing global collaboration.
|
||||
|
||||
Official website express: [https://flagos.io](https://flagos.io/)
|
||||
|
||||
<details>
|
||||
<summary>FlagOS multi-chip support and usage</summary>
|
||||
|
||||
#### FlagOS: Supporting Multiple AI Chips
|
||||
|
||||
Thanks to FlagOS’s unified multi-chip AI system software stack, MiniCPM5-2B was adapted to 9 different AI chips in an extremely short time. Currently, the multi-chip version of MiniCPM5-2B has been released on FlagRelease, FlagOS’s platform for automatic migration, adaptation, and deployment of large models across multi-architecture AI chips. Details are as follows:
|
||||
|
||||
| Vendor | ModelScope | Huggingface |
|
||||
| --------- | --------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------- |
|
||||
| Nvidia | [MiniCPM5-2B-nvidia-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-nvidia-FlagOS) | [MiniCPM5-2B-nvidia-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-nvidia-FlagOS) |
|
||||
| Hygon | [MiniCPM5-2B-hygon-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-hygon-FlagOS) | [MiniCPM5-2B-hygon-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-hygon-FlagOS) |
|
||||
| Metax | [MiniCPM5-2B-metax-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-metax-FlagOS) | [MiniCPM5-2B-metax-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-metax-FlagOS) |
|
||||
| Iluvatar | [MiniCPM5-2B-iluvatar-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-iluvatar-FlagOS) | [MiniCPM5-2B-iluvatar-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-iluvatar-FlagOS) |
|
||||
| Zhenwu | [MiniCPM5-2B-zhenwu-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-zhenwu-FlagOS) | [MiniCPM5-2B-zhenwu-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-zhenwu-FlagOS) |
|
||||
| Mthreads | [MiniCPM5-2B-mthreads-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-mthreads-FlagOS) | [MiniCPM5-2B-mthreads-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-mthreads-FlagOS) |
|
||||
| Kunlunxin | [MiniCPM5-2B-kunlunxin-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-kunlunxin-FlagOS) | [MiniCPM5-2B-kunlunxin-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-kunlunxin-FlagOS) |
|
||||
| Ascend | [MiniCPM5-2B-ascend-FlagOS](https://modelscope.cn/models/FlagRelease/MiniCPM5-2B-ascend-FlagOS) | [MiniCPM5-2B-ascend-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-ascend-FlagOS) |
|
||||
| ARM-v9 | [MiniCPM5-2B-Armv9-FlagOS](https://modelscope.cn/models/FlagRelease/MiniCPM5-2B-Armv9-FlagOS) | [MiniCPM5-2B-Armv9-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-Armv9-FlagOS) |
|
||||
|
||||
#### FlagOS Usage
|
||||
|
||||
##### FlagOS Performance Acceleration on Nvidia
|
||||
|
||||
###### From FlagRelease (**Recommendation**)
|
||||
|
||||
FlagRelease is a platform developed by the FlagOS team for automatic migration, adaptation, and deployment of large models across multi-architecture AI chips. The multi-chip version of MiniCPM5-2B has already been released on FlagRelease. All necessary software packages are pre-installed on the platform, so users do not need to install anything.
|
||||
|
||||
###### FlagRelease Image Key Versions
|
||||
|
||||
###### FlagRelease Quick Start
|
||||
|
||||
| Vendor | ModelScope | Huggingface |
|
||||
| --------- | --------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------- |
|
||||
| Nvidia | [MiniCPM5-2B-nvidia-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-nvidia-FlagOS) | [MiniCPM5-2B-nvidia-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-nvidia-FlagOS) |
|
||||
| Hygon | [MiniCPM5-2B-hygon-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-hygon-FlagOS) | [MiniCPM5-2B-hygon-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-hygon-FlagOS) |
|
||||
| Metax | [MiniCPM5-2B-metax-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-metax-FlagOS) | [MiniCPM5-2B-metax-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-metax-FlagOS) |
|
||||
| Iluvatar | [MiniCPM5-2B-iluvatar-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-iluvatar-FlagOS) | [MiniCPM5-2B-iluvatar-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-iluvatar-FlagOS) |
|
||||
| Zhenwu | [MiniCPM5-2B-zhenwu-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-zhenwu-FlagOS) | [MiniCPM5-2B-zhenwu-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-zhenwu-FlagOS) |
|
||||
| Mthreads | [MiniCPM5-2B-mthreads-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-mthreads-FlagOS) | [MiniCPM5-2B-mthreads-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-mthreads-FlagOS) |
|
||||
| Kunlunxin | [MiniCPM5-2B-kunlunxin-FlagOS](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-kunlunxin-FlagOS) | [MiniCPM5-2B-kunlunxin-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-kunlunxin-FlagOS) |
|
||||
| Ascend | [MiniCPM5-2B-ascend-FlagOS](https://modelscope.cn/models/FlagRelease/MiniCPM5-2B-ascend-FlagOS) | [MiniCPM5-2B-ascend-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-ascend-FlagOS) |
|
||||
| ARM-v9 | [MiniCPM5-2B-Armv9-FlagOS](https://modelscope.cn/models/FlagRelease/MiniCPM5-2B-Armv9-FlagOS) | [MiniCPM5-2B-Armv9-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-Armv9-FlagOS) |
|
||||
|
||||
###### From Scratch
|
||||
|
||||
- Dependencies: Python 3.12, GLIBC 2.39, GLIBCXX 3.4.33, CXXABI 1.3.15
|
||||
|
||||
###### Vllm Version
|
||||
|
||||
###### Installing the FlagOS Operator Library
|
||||
|
||||
Official Repository: https://github.com/flagos-ai/FlagGems
|
||||
|
||||
```PowerShell
|
||||
pip install flag-gems==4.2.1rc0
|
||||
pip install triton==3.5.1
|
||||
```
|
||||
|
||||
###### Activating Acceleration
|
||||
|
||||
You can enable flagGems acceleration by adding the import of flagGems in the source code of vllm where inference is performed.
|
||||
|
||||
```Bash
|
||||
import flag_gems
|
||||
flag_gems.enable(record=True, once=True, path="/root/gems.txt")
|
||||
```
|
||||
|
||||
```PowerShell
|
||||
vllm serve ${model_path} \
|
||||
--trust-remote-code \
|
||||
--dtype bfloat16 \
|
||||
--enforce-eager \
|
||||
--port ${Port} \
|
||||
--served-model-name ${model_name} \
|
||||
--gpu-memory-utilization 0.85
|
||||
```
|
||||
|
||||
##### Using FlagOS Unified Multi-Chip Backend Plugin
|
||||
|
||||
[**vllm-plugin-FL**](https://github.com/flagos-ai/vllm-plugin-FL) is a plugin built for the vLLM inference/service framework. Developed on top of FlagOS’s unified multi-chip backend, it is designed to extend vLLM’s capabilities and performance across a variety of hardware environments.
|
||||
|
||||
###### Using vllm-plugin-FL
|
||||
|
||||
| Vendor | From Scratch | From FlagRelease | |
|
||||
| ------ | -------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------ | ----------------------------------------------------------------------------------------- |
|
||||
| Nvidia | [vllm-plugin-FL/MiniCPM5-2B](https://github.com/flagos-ai/vllm-plugin-FL/blob/main/examples/minicpm/README.md) | [MiniCPM5-2B-ModelScope](https://www.modelscope.cn/models/FlagRelease/MiniCPM5-2B-nvidia-FlagOS) | [MiniCPM5-2B-nvidia-FlagOS](https://huggingface.co/FlagRelease/MiniCPM5-2B-nvidia-FlagOS) |
|
||||
|
||||
</details>
|
||||
|
||||
## Limitations and Disclaimer
|
||||
|
||||
This model has no autonomous intent or legal personhood; its outputs are text generated from statistical patterns and may be inaccurate, biased, or offensive, and may be manipulated by carefully crafted prompts ("jailbreaks") into producing unintended content. Its responses on sensitive topics such as politics, health, finance, and law are not reviewed by experts and should not be treated as professional advice.
|
||||
|
||||
This model is provided "**AS IS**", without warranty of any kind, express or implied, and the developers are not liable for any damages arising from its use. Users must employ the model only for lawful, compliant, and ethical purposes, configure their own safeguards, and label AI-generated content where required; deliberate jailbreaking, injection attacks, or inducing harmful output is prohibited, and any such testing is at the user's own risk.
|
||||
|
||||
## License
|
||||
|
||||
This repository and MiniCPM model weights are released under the [Apache-2.0](https://github.com/OpenBMB/MiniCPM/blob/main/LICENSE) License.
|
||||
|
||||
## Citation
|
||||
|
||||
Please cite our paper if you find our work valuable:
|
||||
|
||||
```bibtex
|
||||
@article{minicpm4,
|
||||
title={Minicpm4: Ultra-efficient llms on end devices},
|
||||
author={MiniCPM, Team},
|
||||
journal={arXiv preprint arXiv:2506.07900},
|
||||
year={2025}
|
||||
}
|
||||
```
|
||||
177
chat_template.jinja
Normal file
177
chat_template.jinja
Normal file
@@ -0,0 +1,177 @@
|
||||
{{- bos_token }}{%- if tools %}
|
||||
{%- set tool_definitions %}
|
||||
{{- "# Tools\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
||||
{%- for tool in tools %}
|
||||
{{- "\n" }}
|
||||
{{- tool | tojson(ensure_ascii=False) }}
|
||||
{%- endfor %}
|
||||
{{- '\n</tools>\n\nTool usage guidelines:\n- You may call zero or more functions. If no function calls are needed, just answer normally and do not include any <function ... </function>.\n- When calling a function, return an XML object within <function ... </function> using:\n<function name="function-name"><param name="param-name">param-value</param></function>\n- param-value may be multi-line. If it contains <, & or newline characters, wrap it in a CDATA block: <param name="param-name"><![CDATA[...multi-line value...]]></param>' }}
|
||||
{%- endset %}
|
||||
|
||||
{{- '<|im_start|>system\n' }}
|
||||
{%- if messages[0].role == 'system' %}
|
||||
{%- if '<tool_def_sep>' in messages[0].content %}
|
||||
{{- messages[0].content.replace('<tool_def_sep>', tool_definitions) }}
|
||||
{%- else %}
|
||||
{{- messages[0].content + '\n\n' + tool_definitions }}
|
||||
{%- endif %}
|
||||
{%- else %}
|
||||
{{- tool_definitions.lstrip() }}
|
||||
{%- endif %}
|
||||
{{- '<|im_end|>\n' }}
|
||||
{%- else %}
|
||||
{%- if messages[0].role == 'system' %}
|
||||
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
||||
{%- for message in messages[::-1] %}
|
||||
{%- set index = (messages|length - 1) - loop.index0 %}
|
||||
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
||||
{%- set ns.multi_step_tool = false %}
|
||||
{%- set ns.last_query_index = index %}
|
||||
{%- endif %}
|
||||
{%- endfor %}
|
||||
{%- for message in messages %}
|
||||
{%- if message.content is string %}
|
||||
{%- set content = message.content %}
|
||||
{%- else %}
|
||||
{%- set content = '' %}
|
||||
{%- endif %}
|
||||
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
||||
{%- elif message.role == "assistant" %}
|
||||
{%- set reasoning_content = '' %}
|
||||
{%- if message.reasoning_content is string %}
|
||||
{%- set reasoning_content = message.reasoning_content %}
|
||||
{%- else %}
|
||||
{%- if '</think>' in content %}
|
||||
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
||||
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
|
||||
{%- if message.tool_calls %}
|
||||
{%- set content_parts = content.split('<tool_sep>') %}
|
||||
{%- set processed_content = content_parts[0] %}
|
||||
{%- set tool_calls_count = message.tool_calls|length %}
|
||||
{%- set tool_sep_count = content_parts|length - 1 %}
|
||||
{%- set min_count = [tool_calls_count, tool_sep_count]|min %}
|
||||
|
||||
{%- for i in range(1, content_parts|length) %}
|
||||
{%- set tool_index = i - 1 %}
|
||||
{%- if tool_index < tool_calls_count %}
|
||||
{%- set tool_call = message.tool_calls[tool_index] %}
|
||||
{%- if tool_call.function %}
|
||||
{%- set tool_call = tool_call.function %}
|
||||
{%- endif %}
|
||||
{%- set single_tool_xml %}
|
||||
{{- '<function name="' ~ tool_call.name ~ '">' }}
|
||||
{%- if tool_call.arguments %}
|
||||
{%- set args_dict = tool_call.arguments %}
|
||||
{%- for param_name, param_value in args_dict.items() %}
|
||||
{{- '<param name="' ~ param_name ~ '">' }}
|
||||
{%- if param_value is string and ('<' in param_value or '&' in param_value or '\n' in param_value) %}
|
||||
{{- '<![CDATA[' + param_value + ']]>' }}
|
||||
{%- else %}
|
||||
{{- param_value }}
|
||||
{%- endif %}
|
||||
{{- '</param>' }}
|
||||
{%- endfor %}
|
||||
{%- endif %}
|
||||
{{- '</function>' }}
|
||||
{%- endset %}
|
||||
{%- set processed_content = processed_content + single_tool_xml + content_parts[i] %}
|
||||
{%- else %}
|
||||
{%- set processed_content = processed_content + content_parts[i] %}
|
||||
{%- endif %}
|
||||
{%- endfor %}
|
||||
|
||||
{%- if tool_calls_count > tool_sep_count %}
|
||||
{%- for remaining_index in range(tool_sep_count, tool_calls_count) %}
|
||||
{%- set tool_call = message.tool_calls[remaining_index] %}
|
||||
{%- if tool_call.function %}
|
||||
{%- set tool_call = tool_call.function %}
|
||||
{%- endif %}
|
||||
{%- set remaining_tool_xml %}
|
||||
{{- '<function name="' ~ tool_call.name ~ '">' }}
|
||||
{%- if tool_call.arguments %}
|
||||
{%- set args_dict = tool_call.arguments %}
|
||||
{%- for param_name, param_value in args_dict.items() %}
|
||||
{{- '<param name="' ~ param_name ~ '">' }}
|
||||
{%- if param_value is string and ('<' in param_value or '&' in param_value or '\n' in param_value) %}
|
||||
{{- '<![CDATA[' + param_value + ']]>' }}
|
||||
{%- else %}
|
||||
{{- param_value }}
|
||||
{%- endif %}
|
||||
{{- '</param>' }}
|
||||
{%- endfor %}
|
||||
{%- endif %}
|
||||
{{- '</function>' }}
|
||||
{%- endset %}
|
||||
{%- set processed_content = processed_content + remaining_tool_xml %}
|
||||
{%- endfor %}
|
||||
{%- endif %}
|
||||
|
||||
{%- set content = processed_content %}
|
||||
{%- endif %}
|
||||
|
||||
{%- if reasoning_content %}
|
||||
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
||||
{%- elif '<think>' not in content and '</think>' not in content %}
|
||||
{{- '<|im_start|>' + message.role + '\n<think>\n\n</think>\n\n' + content.lstrip('\n') }}
|
||||
{%- else %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||
{%- endif %}
|
||||
|
||||
{%- if message.tool_calls and not has_tool_sep %}
|
||||
{%- for tool_call in message.tool_calls %}
|
||||
{%- if (loop.first and content) or (not loop.first) %}
|
||||
{{- '\n' }}
|
||||
{%- endif %}
|
||||
{%- if tool_call.function %}
|
||||
{%- set tool_call = tool_call.function %}
|
||||
{%- endif %}
|
||||
{{- '<function name="' ~ tool_call.name ~ '">' }}
|
||||
{%- if tool_call.arguments %}
|
||||
{%- set args_dict = tool_call.arguments %}
|
||||
{%- for param_name, param_value in args_dict.items() %}
|
||||
{{- '<param name="' ~ param_name ~ '">' }}
|
||||
{%- if param_value is string and ('<' in param_value or '&' in param_value or '\n' in param_value) %}
|
||||
{{- '<![CDATA[' + param_value + ']]>' }}
|
||||
{%- else %}
|
||||
{{- param_value }}
|
||||
{%- endif %}
|
||||
{{- '</param>' }}
|
||||
{%- endfor %}
|
||||
{%- endif %}
|
||||
{{- '</function>' }}
|
||||
{%- endfor %}
|
||||
{%- endif %}
|
||||
{{- '<|im_end|>\n' }}
|
||||
{%- elif message.role == "tool" %}
|
||||
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
||||
{{- '<|im_start|>user' }}
|
||||
{%- endif %}
|
||||
{{- '\n<tool_response>\n' }}
|
||||
{%- if message.content is string %}
|
||||
{{- content }}
|
||||
{%- else %}
|
||||
{{- message.content | tojson(ensure_ascii=False) }}
|
||||
{%- endif %}
|
||||
{{- '\n</tool_response>' }}
|
||||
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
||||
{{- '<|im_end|>\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- endfor %}
|
||||
{%- if add_generation_prompt %}
|
||||
{{- '<|im_start|>assistant\n' }}
|
||||
{%- if enable_thinking is defined %}
|
||||
{%- if enable_thinking is false %}
|
||||
{{- '<think>\n\n</think>\n\n' }}
|
||||
{%- elif enable_thinking is true %}
|
||||
{{- '<think>\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
27
config.json
Normal file
27
config.json
Normal file
@@ -0,0 +1,27 @@
|
||||
{
|
||||
"_name_or_path": "openbmb/MiniCPM5-2B",
|
||||
"architectures": [
|
||||
"LlamaForCausalLM"
|
||||
],
|
||||
"bos_token_id": 0,
|
||||
"eos_token_id": [1, 130073],
|
||||
"pad_token_id": 1,
|
||||
"hidden_act": "silu",
|
||||
"hidden_size": 2048,
|
||||
"initializer_range": 0.02,
|
||||
"intermediate_size": 6144,
|
||||
"max_position_embeddings": 524288,
|
||||
"model_type": "llama",
|
||||
"num_attention_heads": 16,
|
||||
"num_hidden_layers": 42,
|
||||
"num_key_value_heads": 2,
|
||||
"head_dim": 128,
|
||||
"rms_norm_eps": 1e-06,
|
||||
"rope_theta": 5000000,
|
||||
"rope_scaling": null,
|
||||
"tie_word_embeddings": false,
|
||||
"torch_dtype": "bfloat16",
|
||||
"transformers_version": "5.6.2",
|
||||
"use_cache": true,
|
||||
"vocab_size": 130560
|
||||
}
|
||||
1
configuration.json
Normal file
1
configuration.json
Normal file
@@ -0,0 +1 @@
|
||||
{"framework": "pytorch", "task": "text-generation", "allow_remote": true}
|
||||
13
generation_config.json
Normal file
13
generation_config.json
Normal file
@@ -0,0 +1,13 @@
|
||||
{
|
||||
"_from_model_config": true,
|
||||
"bos_token_id": 0,
|
||||
"eos_token_id": [
|
||||
1,
|
||||
130073
|
||||
],
|
||||
"pad_token_id": 1,
|
||||
"do_sample": true,
|
||||
"temperature": 1.0,
|
||||
"top_p": 0.95,
|
||||
"transformers_version": "5.6.2"
|
||||
}
|
||||
3
model.safetensors
Normal file
3
model.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:d80717e7b8eb21ef43070244ecebd85d6694e4a33602fdb817f366bdb04e1e5a
|
||||
size 5033557128
|
||||
30
special_tokens_map.json
Normal file
30
special_tokens_map.json
Normal file
@@ -0,0 +1,30 @@
|
||||
{
|
||||
"bos_token": {
|
||||
"content": "<s>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false
|
||||
},
|
||||
"eos_token": {
|
||||
"content": "</s>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false
|
||||
},
|
||||
"pad_token": {
|
||||
"content": "</s>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false
|
||||
},
|
||||
"unk_token": {
|
||||
"content": "<unk>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false
|
||||
}
|
||||
}
|
||||
3
tokenizer.json
Normal file
3
tokenizer.json
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:3e065a558a034185fe299917b398685c1facd0169a9eea1e629eb30c171fed81
|
||||
size 9894271
|
||||
4099
tokenizer_config.json
Normal file
4099
tokenizer_config.json
Normal file
File diff suppressed because it is too large
Load Diff
Reference in New Issue
Block a user