2025-11-04 16:59:16 +08:00
|
|
|
|
# enginex-hygon-vl
|
|
|
|
|
|
|
2025-11-26 18:39:06 +08:00
|
|
|
|
运行于【海光 DCU】系列算力卡的【视觉多模态理解】引擎，基于 transformer 架构，支持 MiniCPM-V、Gemma 3 等最新流行模型
|
|
|
|
|
|
|
|
|
|
|
|
因具体模型之间的启动方式和具体镜像会有略微差别，请详细查看 `/enginex` 目录下各个支持模型的启动测试方式。
|
|
|
|
|
|
|
|
|
|
|
|
## 可支持模型列表
|
|
|
|
|
|
可在项目文件夹 `/enginex` 下查看具体可支持模型文件的运行方式。
|
|
|
|
|
|
|
|
|
|
|
|
支持模型列表：
|
2025-11-27 01:53:53 +08:00
|
|
|
|
- deepseek-ai/Janus-Pro-7B
|
2025-11-26 18:39:06 +08:00
|
|
|
|
- PaddlePaddle/PaddleOCR-VL
|
|
|
|
|
|
- Qwen/Qwen2.5-7B-instruct
|
2025-11-27 01:53:53 +08:00
|
|
|
|
- Qwen/Qwen2-VL-2B-Instruct
|
2025-11-27 02:53:34 +08:00
|
|
|
|
- ZhipuAI/glm-4v-9b
|
2025-12-01 18:45:45 +08:00
|
|
|
|
- LLM-Research/gemma-3-27b-it
|
2025-12-04 20:17:51 +08:00
|
|
|
|
- stepfun-ai/step3
|
|
|
|
|
|
- OpenBMB/MiniCPM-V
|
|
|
|
|
|
- huangjianuo/llava-v1.5-7b
|
|
|
|
|
|
- gpt-omni/mini-omni2
|
|
|
|
|
|
- FoundationVision/var
|
|
|
|
|
|
- AIDC-AI/Ovis1.6-Gemma2-9B
|
|
|
|
|
|
- VITA-MLLM/VITA-1.5
|
|
|
|
|
|
- AIDC-AI/Marco-o1
|
|
|
|
|
|
- moonshotai/Kimi-VL-A3B-Thinking
|
|
|
|
|
|
- AI-ModelScope/Florence-2-large-ft
|
|
|
|
|
|
- BLIP3o/BLIP3o-Model
|
|
|
|
|
|
- AI-ModelScope/OmniParser
|
2025-12-05 11:01:32 +08:00
|
|
|
|
- moonshotai/Kimi-VL-A3B-Instruct
|
|
|
|
|
|
- wlc952/mineru
|
|
|
|
|
|
- ByteDance-Seed/UI-TARS-1.5-7B
|
