whatmodel.sh / BioMistral-7B
BioMistral-7B
↗ share
+ contribute
BioMistral-7B is an open-weights language model from BioMistral, released February 2024. Licence: apache-2.0.
Catalogue identity notes
Maker repo BioMistral/BioMistral-7B @9a11e1f: licence tag apache-2.0. Released 2024-02: month the maker's Hugging Face repo was created (createdAt 2024-02-14).
$ how it compares
The LLMs ranked either side of BioMistral-7B, out of 828 with a score.
| # | model | score | where |
|---|---|---|---|
| 657 |  Granite 3B Code Instruct (2K) | 38±2 | local |
| 658 |  DeepSeek-V2 Lite Chat (16B-A2.4B) | 38±1 | local |
| 659 | BioMistral-7B | 38±3 | local |
| 660 |  Qwen3.5 0.8B | 38±1 | local |
| 661 |  Code Tulu 2 7B | 38±2 | local |
$ what it's good at
other tests3 not in a task · 95% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| BRIDGE clinical text, zero-shot average (87 tasks) | 20.4% ✎ | ▁▁▁▁▁▃▂▃▄▆█▃ beats 12/97 | 39% |
| Polish medical board exams, bias-mitigated MCQ protocol (7-exam average) | 30.6% ✎ | ▄·▂▂▂··█▂▄·▄ beats 1/13 | 39% |
| SurgiQ surgical MCQ (zero-shot, log-probability scoring) | 47.1% ✎ | ▂···▂▂·▃▃▃██ beats 5/17 | 17% |
$ where to run it
llama.cpp · recommended
version
- Install llama.cpp · Homebrew and winget installs name the commands llama-cli and llama-server instead.macOS / Linux
curl -LsSf https://llama.app/install.sh | shWindowsirm https://llama.app/install.ps1 | iex - Run BioMistral-7B · needs ~4.4 GB · from huggingface.co/BioMistral (official)chat in the terminal
llama cli -hf BioMistral/BioMistral-7B-GGUF:Q4_K_M - Or start a local server and open the address it prints (http://localhost:9931)
llama serve -hf BioMistral/BioMistral-7B-GGUF:Q4_K_M
LM Studio
version
- Install LM Studio · Open the app once before using the lms command.
- Run BioMistral-7B · needs ~4.4 GB · from huggingface.co/BioMistral (official)
- Or download it in the terminal
lms get https://huggingface.co/BioMistral/BioMistral-7B-GGUF@Q4_K_M -y - Then run lms ls to see its name, and lms chat followed by that name
Ollama
version
- Install Ollama
download it (macOS / Windows) ↗
Linuxcurl -fsSL https://ollama.com/install.sh | sh - Run BioMistral-7B · needs ~4.4 GB · from huggingface.co/BioMistral (official)chat in the terminal
ollama run hf.co/BioMistral/BioMistral-7B-GGUF:Q4_K_M
Other ways to run it
- Run it as published · needs ~14.48 GB · from huggingface.co/BioMistral (official)
uv run --with torch --with 'transformers<5' --with accelerate python -c "from transformers import pipeline; print(pipeline('text-generation', model='BioMistral/BioMistral-7B', torch_dtype='auto', device_map='auto')('Once upon a time', max_new_tokens=40)[0]['generated_text'])"
Official weights and documentation: huggingface.co/BioMistral ↗
$ how the score was worked out · BRIDGE Medical Leaderboard (H… 39%, Reassessing High-Performing L… 39%, SurgiQ: A Large-Scale Multi-D… 17%
Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works
| source | results | share of the score |
|---|---|---|
 BRIDGE Medical Leaderboard (Harvard Medical School / Brigham and Women's YLab) | 1 | 39% |
 Reassessing High-Performing LLMs on Polish Medical Exams: True Competence or Bias-Driven Performance? (arXiv:2606.12250v1) | 1 | 39% |
| SurgiQ: A Large-Scale Multi-Domain Benchmark for Evaluating Surgical Understanding in Large Language Models (arXiv:2606.08071v1) | 1 | 17% |
| prior · pull toward the middle | 5% | |
| weighted average | = 38.0 |
$ improve this page
Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.
data updated · sources and licences · json