MiaoMint/Ternary-Bonsai-1.7B-gguf
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

初始化项目，由ModelHub XC社区提供模型

Browse Source 
Model: MiaoMint/Ternary-Bonsai-1.7B-gguf
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-06-05 17:19:18 +08:00
commit d0bcc74182
10 changed files with 582 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
.eval_results/gsm8k.yaml Normal file
View File

@@ -0,0 +1,7 @@
- dataset:
id: openai/gsm8k
task_id: gsm8k
value: 74.2
source:
url: https://huggingface.co/prism-ml/Bonsai-8B-gguf
name: Model Card

40
.gitattributes vendored Normal file
View File

@@ -0,0 +1,40 @@
*.7z filter=lfs diff=lfs merge=lfs -text
*.arrow filter=lfs diff=lfs merge=lfs -text
*.bin filter=lfs diff=lfs merge=lfs -text
*.bz2 filter=lfs diff=lfs merge=lfs -text
*.ckpt 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
*.mlmodel filter=lfs diff=lfs merge=lfs -text
*.model filter=lfs diff=lfs merge=lfs -text
*.msgpack filter=lfs diff=lfs merge=lfs -text
*.npy filter=lfs diff=lfs merge=lfs -text
*.npz 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
*.pickle filter=lfs diff=lfs merge=lfs -text
*.pkl 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
*.safetensors filter=lfs diff=lfs merge=lfs -text
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
*.tar.* 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
*.wasm filter=lfs diff=lfs merge=lfs -text
*.xz filter=lfs diff=lfs merge=lfs -text
*.zip 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
Ternary-Bonsai-1.7B-F16.gguf filter=lfs diff=lfs merge=lfs -text
Ternary-Bonsai-1.7B-Q2_0.gguf filter=lfs diff=lfs merge=lfs -text
Ternary-Bonsai-1.7B-TQ1_0.gguf filter=lfs diff=lfs merge=lfs -text
Ternary-Bonsai-1.7B-TQ2_0.gguf filter=lfs diff=lfs merge=lfs -text

177
LICENSE Normal file
View File

@@ -0,0 +1,177 @@
Apache License
Version 2.0, January 2004
http://www.apache.org/licenses/
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
1. Definitions.
"License" shall mean the terms and conditions for use, reproduction,
and distribution as defined by Sections 1 through 9 of this document.
"Licensor" shall mean the copyright owner or entity authorized by
the copyright owner that is granting the License.
"Legal Entity" shall mean the union of the acting entity and all
other entities that control, are controlled by, or are under common
control with that entity. For the purposes of this definition,
"control" means (i) the power, direct or indirect, to cause the
direction or management of such entity, whether by contract or
otherwise, or (ii) ownership of fifty percent (50%) or more of the
outstanding shares, or (iii) beneficial ownership of such entity.
"You" (or "Your") shall mean an individual or Legal Entity
exercising permissions granted by this License.
"Source" form shall mean the preferred form for making modifications,
including but not limited to software source code, documentation
source, and configuration files.
"Object" form shall mean any form resulting from mechanical
transformation or translation of a Source form, including but
not limited to compiled object code, generated documentation,
and conversions to other media types.
"Work" shall mean the work of authorship, whether in Source or
Object form, made available under the License, as indicated by a
copyright notice that is included in or attached to the work
(an example is provided in the Appendix below).
"Derivative Works" shall mean any work, whether in Source or Object
form, that is based on (or derived from) the Work and for which the
editorial revisions, annotations, elaborations, or other modifications
represent, as a whole, an original work of authorship. For the purposes
of this License, Derivative Works shall not include works that remain
separable from, or merely link (or bind by name) to the interfaces of,
the Work and Derivative Works thereof.
"Contribution" shall mean any work of authorship, including
the original version of the Work and any modifications or additions
to that Work or Derivative Works thereof, that is intentionally
submitted to Licensor for inclusion in the Work by the copyright owner
or by an individual or Legal Entity authorized to submit on behalf of
the copyright owner. For the purposes of this definition, "submitted"
means any form of electronic, verbal, or written communication sent
to the Licensor or its representatives, including but not limited to
communication on electronic mailing lists, source code control systems,
and issue tracking systems that are managed by, or on behalf of, the
Licensor for the purpose of discussing and improving the Work, but
excluding communication that is conspicuously marked or otherwise
designated in writing by the copyright owner as "Not a Contribution."
"Contributor" shall mean Licensor and any individual or Legal Entity
on behalf of whom a Contribution has been received by Licensor and
subsequently incorporated within the Work.
2. Grant of Copyright License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
copyright license to reproduce, prepare Derivative Works of,
publicly display, publicly perform, sublicense, and distribute the
Work and such Derivative Works in Source or Object form.
3. Grant of Patent License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
(except as stated in this section) patent license to make, have made,
use, offer to sell, sell, import, and otherwise transfer the Work,
where such license applies only to those patent claims licensable
by such Contributor that are necessarily infringed by their
Contribution(s) alone or by combination of their Contribution(s)
with the Work to which such Contribution(s) was submitted. If You
institute patent litigation against any entity (including a
cross-claim or counterclaim in a lawsuit) alleging that the Work
or a Contribution incorporated within the Work constitutes direct
or contributory patent infringement, then any patent licenses
granted to You under this License for that Work shall terminate
as of the date such litigation is filed.
4. Redistribution. You may reproduce and distribute copies of the
Work or Derivative Works thereof in any medium, with or without
modifications, and in Source or Object form, provided that You
meet the following conditions:
(a) You must give any other recipients of the Work or
Derivative Works a copy of this License; and
(b) You must cause any modified files to carry prominent notices
stating that You changed the files; and
(c) You must retain, in the Source form of any Derivative Works
that You distribute, all copyright, patent, trademark, and
attribution notices from the Source form of the Work,
excluding those notices that do not pertain to any part of
the Derivative Works; and
(d) If the Work includes a "NOTICE" text file as part of its
distribution, then any Derivative Works that You distribute must
include a readable copy of the attribution notices contained
within such NOTICE file, excluding those notices that do not
pertain to any part of the Derivative Works, in at least one
of the following places: within a NOTICE text file distributed
as part of the Derivative Works; within the Source form or
documentation, if provided along with the Derivative Works; or,
within a display generated by the Derivative Works, if and
wherever such third-party notices normally appear. The contents
of the NOTICE file are for informational purposes only and
do not modify the License. You may add Your own attribution
notices within Derivative Works that You distribute, alongside
or as an addendum to the NOTICE text from the Work, provided
that such additional attribution notices cannot be construed
as modifying the License.
You may add Your own copyright statement to Your modifications and
may provide additional or different license terms and conditions
for use, reproduction, or distribution of Your modifications, or
for any such Derivative Works as a whole, provided Your use,
reproduction, and distribution of the Work otherwise complies with
the conditions stated in this License.
5. Submission of Contributions. Unless You explicitly state otherwise,
any Contribution intentionally submitted for inclusion in the Work
by You to the Licensor shall be under the terms and conditions of
this License, without any additional terms or conditions.
Notwithstanding the above, nothing herein shall supersede or modify
the terms of any separate license agreement you may have executed
with Licensor regarding such Contributions.
6. Trademarks. This License does not grant permission to use the trade
names, trademarks, service marks, or product names of the Licensor,
except as required for reasonable and customary use in describing the
origin of the Work and reproducing the content of the NOTICE file.
7. Disclaimer of Warranty. Unless required by applicable law or
agreed to in writing, Licensor provides the Work (and each
Contributor provides its Contributions) on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied, including, without limitation, any warranties or conditions
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
PARTICULAR PURPOSE. You are solely responsible for determining the
appropriateness of using or redistributing the Work and assume any
risks associated with Your exercise of permissions under this License.
8. Limitation of Liability. In no event and under no legal theory,
whether in tort (including negligence), contract, or otherwise,
unless required by applicable law (such as deliberate and grossly
negligent acts) or agreed to in writing, shall any Contributor be
liable to You for damages, including any direct, indirect, special,
incidental, or consequential damages of any character arising as a
result of this License or out of the use or inability to use the
Work (including but not limited to damages for loss of goodwill,
work stoppage, computer failure or malfunction, or any and all
other commercial damages or losses), even if such Contributor
has been advised of the possibility of such damages.
9. Accepting Warranty or Additional Liability. While redistributing
the Work or Derivative Works thereof, You may choose to offer,
and charge a fee for, acceptance of support, warranty, indemnity,
or other liability obligations and/or rights consistent with this
License. However, in accepting such obligations, You may act only
on Your own behalf and on Your sole responsibility, not on behalf
of any other Contributor, and only if You agree to indemnify,
defend, and hold each Contributor harmless for any liability
incurred by, or claims asserted against, such Contributor by reason
of your accepting any such warranty or additional liability.
END OF TERMS AND CONDITIONS

4
NOTICE.txt Normal file
View File

@@ -0,0 +1,4 @@
This software is copyright 2026-present Prism ML, Inc. It is available under the Apache 2.0 license.
If you publicly deploy or redistribute this software, we would appreciate attribution such as: “Created using Bonsai by Prism ML."
This software is built from Qwen3-1.7B, Copyright 2024 Alibaba Cloud, which is available under the Apache 2.0 License: https://huggingface.co/Qwen/Qwen3-1.7B/blob/main/LICENSE

175
README.md Normal file
View File

@@ -0,0 +1,175 @@
---
license: apache-2.0
library_name: gguf
pipeline_tag: text-generation
tags:
- ternary
- 1.58-bit
- gguf
- llama-cpp
- q2_0
- on-device
- prismml
- bonsai
base_model:
- prism-ml/Ternary-Bonsai-1.7B-unpacked
---
<p align="center">
<img src="./assets/bonsai-logo.svg" width="280" alt="Bonsai">
</p>
<p align="center">
<a href="https://prismml.com"><b>Prism ML Website</b></a> &nbsp;|&nbsp;
<a href="https://github.com/PrismML-Eng/Bonsai-demo/blob/main/ternary-bonsai-8b-whitepaper.pdf"><b>White Paper</b></a> &nbsp;|&nbsp;
<a href="https://github.com/PrismML-Eng/Bonsai-demo"><b>Demo &amp; Examples</b></a> &nbsp;|&nbsp;
<a href="https://discord.gg/prismml"><b>Discord</b></a>
</p>
# Ternary-Bonsai-1.7B-gguf
Ternary (1.58-bit) language model in GGUF Q2_0 format for `llama.cpp`
<p align="center">
<img src="./assets/frontier.svg" width="680" alt="Pareto Frontier">
</p>
## Resources
- **[White Paper](https://github.com/PrismML-Eng/Bonsai-demo/blob/main/ternary-bonsai-8b-whitepaper.pdf)**
- **[Demo repo](https://github.com/PrismML-Eng/Bonsai-demo)** — examples for serving, benchmarking, and integrating Bonsai
- **[Discord](https://discord.gg/prismml)** — community support and updates
- **Kernels**: Q2_0 is not yet in mainline `llama.cpp`. Use our fork at [PrismML-Eng/llama.cpp](https://github.com/PrismML-Eng/llama.cpp) (`prism` branch, default) which adds Q2_0 support for CPU (NEON/generic) and Metal. Upstream PR coming soon.
## Model Overview
| Item | Specification |
| :--------------- | :----------------------------------------------------------------------- |
| Base model | Qwen3-1.7B |
| Parameters | 1.72B |
| Architecture | GQA, SwiGLU MLP, RoPE, RMSNorm |
| Context length | 32,768 tokens |
| Vocab size | 151,936 |
| Weight format | GGUF Q2_0 g128: {-1, 0, +1} with FP16 group-wise scaling |
| Packed Q2_0 size | **436 MiB** (0.46 GB) |
| Ternary coverage | Embeddings, attention projections, MLP projections, LM head |
| License | Apache 2.0 |
## Quantization Format: GGUF Q2_0 (g128)
Each weight takes a value from {-1, 0, +1}, with one shared FP16 scale per group of 128 weights:
```
w_i = scale_g * t_i, t_i in {-1, 0, +1}
```
Q2_0 encodes each weight as a 2-bit code `q in {0, 1, 2, 3}`, dequantized via `w = (q - 1) * scale`. One 128-element block is 34 bytes (2 bytes FP16 scale + 32 bytes of packed 2-bit codes) for an effective **2.125 bits/weight**. The fourth code point (`q = 3`, reconstructing to `+2 * scale`) is reserved for future extensions; for ternary weights it is unused.
### Memory
| Format | Size | Reduction | Ratio |
| :---------------- | ----------: | --------: | ---------: |
| FP16 | 3.44 GB | -- | 1.0x |
| **GGUF Q2_0 g128**| **436 MiB** (0.46 GB) | **86.6%** | **7.5x** |
## Files in this repo
| File | Format | Size | Recommended |
| :-------------------------------- | :----- | -----: | :---------- |
| `Ternary-Bonsai-1.7B-F16.gguf` | FP16 | 3.44 GB | baseline / re-quantization source |
| `Ternary-Bonsai-1.7B-Q2_0.gguf` | Q2_0 (g128) | 442 MB | **recommended** (lossless for ternary) |
## Quickstart
### Build from the Prism fork
```bash
git clone https://github.com/PrismML-Eng/llama.cpp
cd llama.cpp
cmake -B build -DGGML_METAL=ON # or -DGGML_CUDA=ON, -DGGML_VULKAN=ON
cmake --build build -j
```
### `llama.cpp` CLI
```bash
./build/bin/llama-cli \
-m Ternary-Bonsai-1.7B-Q2_0.gguf \
-p "Explain quantum computing in simple terms." \
-n 256
```
### `llama.cpp` server
```bash
./build/bin/llama-server -m Ternary-Bonsai-1.7B-Q2_0.gguf -c 4096
```
## Throughput (llama.cpp, Apple M4 Pro 48 GB)
| Backend | PP512 (tok/s) | TG128 (tok/s) | FP16 TG (tok/s) | Speedup |
| :--------------- | ------------: | ------------: | ---------------: | ---------: |
| Metal (GPU) | 2,088 | **229** | 60 | **3.8x** |
| NEON CPU (10 t) | 508 | **123** | — | — |
Flags: `-ngl 99 -fa 1` for Metal; `-ngl 0 -fa 1 -t 10` for CPU.
## Fidelity (Q2_0 vs FP16 baseline)
Measured on wikitext-2 (20 chunks, context 512) via `llama-perplexity --kl-divergence`:
| Metric | Value |
| :------------------ | ---------: |
| Mean KL | 0.000000 |
| Top-1 agreement | 100.000 % |
| RMS Δp | 0.015 % |
| PPL ratio (Q/base) | 1.0048 |
Q2_0 is effectively lossless for ternary weights — the ternary values land exactly on three of the four 2-bit code points.
## Benchmarks
Evaluated with EvalScope v1.4.2 + vLLM 0.15.1 on NVIDIA H100. Full benchmark suite:
| Model | Size | Avg | MMLU-R | MuSR | IFEval | GSM8K | HE+ | BFCLv3 |
| :-------------------------- | ----------: | --------: | -----: | ---: | -----: | ----: | ---: | -----: |
| **Ternary Bonsai 1.7B** | **0.44 GB** | **58.47** | 52.9 | 50.8 | 70.1 | 74.2 | 51.8 | 51.0 |
| *1-bit Bonsai 1.7B (prior)* | *0.24 GB* | *49.60* | 43.2 | 45.1 | 63.0 | 66.3 | 45.1 | 34.9 |
| Qwen3 1.7B | 3.44 GB | **66.57** | 66.8 | 50.1 | 70.3 | 83.1 | 57.3 | 71.8 |
| Qwen3 0.6B | 1.19 GB | **48.02** | 47.5 | 41.5 | 62.8 | 64.1 | 30.5 | 41.7 |
| LFM2 1.2B | 2.34 GB | **46.73** | 52.9 | 25.4 | 77.5 | 62.2 | 36.0 | 26.4 |
| Gemma3 1B | 2.00 GB | **45.53** | 43.2 | 37.0 | 61.9 | 64.4 | 40.2 | 26.5 |
| Llama 3.2 1B | 2.47 GB | **39.88** | 47.2 | 29.2 | 47.7 | 49.0 | 35.4 | 30.8 |
## Intelligence Density
```
density = -ln(1 - score/100) / size_GB
```
| Model | Size | Intelligence Density (1/GB) |
| :-------------------------- | ----------: | --------------------------: |
| **Ternary Bonsai 1.7B** | **0.44 GB** | **2.001** |
| *1-bit Bonsai 1.7B (prior)* | *0.24 GB* | *2.832* |
| Qwen3 0.6B | 1.19 GB | 0.549 |
| Qwen3 1.7B | 3.44 GB | 0.318 |
| Gemma3 1B | 2.00 GB | 0.304 |
| LFM2 1.2B | 2.34 GB | 0.269 |
| Llama 3.2 1B | 2.47 GB | 0.206 |
## Citation
```bibtex
@techreport{ternarybonsai,
title = {Ternary Bonsai: 1.58-bit Language Models at 8B, 4B, and 1.7B Scale},
author = {Prism ML},
year = {2026},
month = {April},
url = {https://prismml.com}
}
```
## Contact
For questions, feedback, or collaboration inquiries: **contact@prismml.com**

3
Ternary-Bonsai-1.7B-F16.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:00be231b1ba8ab8b45db35d288f897fa9b5836bb6ad41762b759b6bc990b4fea
size 3446249408

3
Ternary-Bonsai-1.7B-Q2_0.gguf Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:d97d94eb564590c9f0300e54d3f87bbbb25a78693d0ade9f6e177973dcb8228a
size 463290464

1
assets/bonsai-logo.svg Normal file
View File

@@ -0,0 +1 @@
<svg width="24" height="15" viewBox="0 0 24 15" fill="none" xmlns="http://www.w3.org/2000/svg"><style>path{fill:#000}@media(prefers-color-scheme:dark){path{fill:#fff}}</style><path d="M15.1304 6.25864H12L14.087 10.4311L6.78261 14.6035H16.1739L20.3478 10.4311L15.1304 6.25864Z"/><path d="M16.7826 6.25866H24V8.34488H19.4022L16.7826 6.25866Z"/><path d="M0 6.25866H10.837L11.8804 8.34488H0V6.25866Z"/><path d="M5.21739 3.12933H20.8696V5.21555H5.21739V3.12933Z"/><path d="M10.4348 0H17.7391V2.08622H10.4348V0Z"/></svg>
Side by Side

After

Width:  |  Height:  |  Size: 515 B

BIN
assets/frontier.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 77 KiB

172
assets/frontier.svg Normal file
View File

@@ -0,0 +1,172 @@
<svg xmlns="http://www.w3.org/2000/svg" width="1500" height="680" viewBox="0 0 1500 680">
<style>
.bg { fill: #ede9e4; }
.title { fill: #2d2a27; font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', Helvetica, Arial, sans-serif; font-weight: 700; font-size: 28px; }
.subtitle { fill: #6a6560; font-family: 'SF Mono', 'Fira Code', 'Fira Mono', 'Roboto Mono', 'Courier New', monospace; font-size: 14px; }
.separator { stroke: #c5c0ba; }
.grid { stroke: #d5d0ca; stroke-width: 0.8; }
.axis { stroke: #2d2a27; stroke-width: 1.2; }
.tick-label { fill: #6a6560; font-family: 'SF Mono', 'Fira Code', 'Fira Mono', 'Roboto Mono', 'Courier New', monospace; font-size: 13px; }
.axis-label { fill: #6a6560; font-family: 'SF Mono', 'Fira Code', 'Fira Mono', 'Roboto Mono', 'Courier New', monospace; font-size: 13px; }
.square-marker { fill: #d4b0b3; fill-opacity: 0.45; stroke: #a07578; stroke-width: 2; }
.bonsai-fill { fill: #2d2a27; }
.bonsai-label { fill: #2d2a27; font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', Helvetica, Arial, sans-serif; font-weight: 700; font-size: 14px; }
.prior-fill { fill: #2d2a27; }
.prior-label { fill: #2d2a27; font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', Helvetica, Arial, sans-serif; font-weight: 700; font-size: 14px; }
.model-label { fill: #8a7072; font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', Helvetica, Arial, sans-serif; font-size: 12px; }
.prior-line { stroke: #c0a8ab; stroke-width: 2; stroke-dasharray: 8 5; fill: none; }
.bonsai-line { stroke: #2d2a27; stroke-width: 2.5; stroke-dasharray: 8 5; fill: none; }
@media (prefers-color-scheme: dark) {
.bg { fill: #1e1c1b; }
.title { fill: #e8e4e0; }
.subtitle { fill: #a09890; }
.separator { stroke: #3a3634; }
.grid { stroke: #2e2b29; }
.axis { stroke: #a09890; }
.tick-label { fill: #a09890; }
.axis-label { fill: #a09890; }
.square-marker { fill: #d4b0b3; fill-opacity: 0.35; stroke: #a07578; }
.bonsai-fill { fill: #e8e4e0; }
.bonsai-label { fill: #e8e4e0; }
.prior-fill { fill: #e8e4e0; }
.prior-label { fill: #e8e4e0; }
.model-label { fill: #b89498; }
.prior-line { stroke: #7a6568; }
.bonsai-line { stroke: #e8e4e0; }
}
</style>
<defs>
<symbol id="bonsai-icon" viewBox="0 0 24 14.6035">
<path d="M15.1304 6.25864H12L14.087 10.4311L6.78261 14.6035H16.1739L20.3478 10.4311L15.1304 6.25864Z"/>
<path d="M16.7826 6.25866H24V8.34488H19.4022L16.7826 6.25866Z"/>
<path d="M0 6.25866H10.837L11.8804 8.34488H0V6.25866Z"/>
<path d="M5.21739 3.12933H20.8696V5.21555H5.21739V3.12933Z"/>
<path d="M10.4348 0H17.7391V2.08622H10.4348V0Z"/>
</symbol>
</defs>
<!-- Background -->
<rect width="1500" height="680" class="bg" />
<!-- Title and subtitle -->
<text x="68" y="48" class="title">Frontier efficiency</text>
<text x="68" y="72" class="subtitle">Average score (IFEval, GSM8K, HumanEval+, BFCL, MuSR, MMLU-Redux)</text>
<!-- Separator -->
<line x1="68" y1="85" x2="1440" y2="85" class="separator" stroke-width="1" />
<!--
pixelX(v) = 130 + (ln(v) + 1.8971) / 5.1160 * 1300
pixelY(v) = 570 - (v - 35) * 9.1
Ternary: (0.37,57.47)→(359,366) (0.86,70.65)→(574,246) (1.75,75.48)→(755,202)
1-bit: (0.25,49.30)→(260,440) (0.575,61.6)→(472,328) (1.15,70.5)→(648,247)
-->
<!-- Vertical grid lines -->
<line x1="260" y1="115" x2="260" y2="570" class="grid" />
<line x1="436" y1="115" x2="436" y2="570" class="grid" />
<line x1="612" y1="115" x2="612" y2="570" class="grid" />
<line x1="788" y1="115" x2="788" y2="570" class="grid" />
<line x1="964" y1="115" x2="964" y2="570" class="grid" />
<line x1="1141" y1="115" x2="1141" y2="570" class="grid" />
<line x1="1317" y1="115" x2="1317" y2="570" class="grid" />
<!-- Axes -->
<line x1="130" y1="570" x2="1430" y2="570" class="axis" />
<line x1="130" y1="115" x2="130" y2="570" class="axis" />
<!-- Y-axis tick labels -->
<text x="118" y="529" text-anchor="end" class="tick-label">40</text>
<text x="118" y="438" text-anchor="end" class="tick-label">50</text>
<text x="118" y="347" text-anchor="end" class="tick-label">60</text>
<text x="118" y="256" text-anchor="end" class="tick-label">70</text>
<text x="118" y="165" text-anchor="end" class="tick-label">80</text>
<!-- Y-axis small ticks -->
<line x1="125" y1="525" x2="130" y2="525" class="axis" />
<line x1="125" y1="434" x2="130" y2="434" class="axis" />
<line x1="125" y1="343" x2="130" y2="343" class="axis" />
<line x1="125" y1="252" x2="130" y2="252" class="axis" />
<line x1="125" y1="161" x2="130" y2="161" class="axis" />
<!-- X-axis tick labels -->
<text x="260" y="595" text-anchor="middle" class="tick-label">0.25 GB</text>
<text x="436" y="595" text-anchor="middle" class="tick-label">0.5 GB</text>
<text x="612" y="595" text-anchor="middle" class="tick-label">1 GB</text>
<text x="788" y="595" text-anchor="middle" class="tick-label">2 GB</text>
<text x="964" y="595" text-anchor="middle" class="tick-label">4 GB</text>
<text x="1141" y="595" text-anchor="middle" class="tick-label">8 GB</text>
<text x="1317" y="595" text-anchor="middle" class="tick-label">16 GB</text>
<!-- X-axis small ticks -->
<line x1="260" y1="570" x2="260" y2="575" class="axis" />
<line x1="436" y1="570" x2="436" y2="575" class="axis" />
<line x1="612" y1="570" x2="612" y2="575" class="axis" />
<line x1="788" y1="570" x2="788" y2="575" class="axis" />
<line x1="964" y1="570" x2="964" y2="575" class="axis" />
<line x1="1141" y1="570" x2="1141" y2="575" class="axis" />
<line x1="1317" y1="570" x2="1317" y2="575" class="axis" />
<!-- Axis labels -->
<text x="730" y="635" text-anchor="middle" class="axis-label">Model size in GB (log scale)</text>
<text x="52" y="343" text-anchor="middle" class="axis-label" transform="rotate(-90, 52, 343)">Average benchmark score</text>
<!-- Prior Pareto frontier curve (conventional models) -->
<path d="M 521 572 L 555 541 L 585 515 L 612 492 L 656 456 L 698 424 L 747 389 L 788 361 L 845 326 L 891 300 L 926 282 L 964 263 L 1021 239 L 1067 221 L 1101 210 L 1124 203 L 1142 198 L 1170 191 L 1197 185 L 1221 180 L 1243 176 L 1264 173 L 1283 170 L 1300 168 L 1317 167 L 1323 166 L 1332 166 L 1347 165 L 1360 164 L 1373 164"
class="prior-line" />
<!-- Bonsai frontier line (single line through all 6 points) -->
<path d="M 155 480 L 780 170" class="bonsai-line" />
<!-- Other model squares -->
<rect x="650" y="446" width="12" height="12" rx="1" class="square-marker" />
<rect x="782" y="468" width="12" height="12" rx="1" class="square-marker" />
<rect x="822" y="457" width="12" height="12" rx="1" class="square-marker" />
<rect x="836" y="520" width="12" height="12" rx="1" class="square-marker" />
<rect x="920" y="277" width="12" height="12" rx="1" class="square-marker" />
<rect x="1079" y="297" width="12" height="12" rx="1" class="square-marker" />
<rect x="1095" y="216" width="12" height="12" rx="1" class="square-marker" />
<rect x="1127" y="265" width="12" height="12" rx="1" class="square-marker" />
<rect x="1136" y="181" width="12" height="12" rx="1" class="square-marker" />
<rect x="1243" y="326" width="12" height="12" rx="1" class="square-marker" />
<rect x="1287" y="237" width="12" height="12" rx="1" class="square-marker" />
<rect x="1298" y="382" width="12" height="12" rx="1" class="square-marker" />
<rect x="1311" y="236" width="12" height="12" rx="1" class="square-marker" />
<rect x="1312" y="272" width="12" height="12" rx="1" class="square-marker" />
<rect x="1312" y="287" width="12" height="12" rx="1" class="square-marker" />
<rect x="1312" y="367" width="12" height="12" rx="1" class="square-marker" />
<rect x="1317" y="161" width="12" height="12" rx="1" class="square-marker" />
<rect x="1320" y="217" width="12" height="12" rx="1" class="square-marker" />
<rect x="1321" y="249" width="12" height="12" rx="1" class="square-marker" />
<rect x="1352" y="285" width="12" height="12" rx="1" class="square-marker" />
<!-- 1-Bit Bonsai icons (faded prior work) -->
<use href="#bonsai-icon" x="249" y="433" width="22" height="14" class="prior-fill" />
<use href="#bonsai-icon" x="461" y="321" width="22" height="14" class="prior-fill" />
<use href="#bonsai-icon" x="637" y="240" width="22" height="14" class="prior-fill" />
<!-- 1-Bit labels (well below their icons, clear of line) -->
<text x="180" y="475" class="prior-label">1-Bit Bonsai 1.7B</text>
<text x="400" y="370" class="prior-label">1-Bit Bonsai 4B</text>
<text x="660" y="282" class="prior-label">1-Bit Bonsai 8B</text>
<!-- Ternary Bonsai icons -->
<use href="#bonsai-icon" x="348" y="359" width="22" height="14" class="bonsai-fill" />
<use href="#bonsai-icon" x="563" y="239" width="22" height="14" class="bonsai-fill" />
<use href="#bonsai-icon" x="744" y="195" width="22" height="14" class="bonsai-fill" />
<!-- Ternary Bonsai labels (well above their icons, clear of line) -->
<text x="235" y="318" class="bonsai-label">Ternary Bonsai 1.7B</text>
<text x="445" y="208" class="bonsai-label">Ternary Bonsai 4B</text>
<text x="655" y="155" class="bonsai-label">Ternary Bonsai 8B</text>
<!-- Other model labels (frontier models only) -->
<text x="670" y="445" class="model-label">Qwen3 0.6B</text>
<text x="940" y="280" class="model-label">Qwen3 1.7B</text>
<text x="1115" y="215" class="model-label">Ministral3 3B</text>
<text x="1110" y="175" class="model-label">Qwen3 4B</text>
<text x="1337" y="160" class="model-label">Qwen3 8B</text>
</svg>
Side by Side

After

Width:  |  Height:  |  Size: 9.5 KiB