whatmodel.sh / CoALM-405B
CoALM-405B
↗ share
+ contribute
CoALM-405B is an open-weights language model from UIUC ConvAI & Oumi (CoALM), released February 2025. Its published size is 405.85B parameters. Licence: cc-by-nc-4.0.
Catalogue identity notes
Maker repo uiuc-convai/CoALM-405B @34d882b: licence tag cc-by-nc-4.0. 405,853,388,800 parameters from the Hugging Face safetensors metadata (dense, so active = total). Released 2025-02: month the maker's Hugging Face repo was created (createdAt 2025-02-04).
$ how it compares
The LLMs ranked either side of CoALM-405B, out of 1132 with a score, then more from UIUC ConvAI & Oumi (CoALM).
| # | model | score | where | ||
|---|---|---|---|---|---|
| 306 |  Amazon Nova Experimental Chat (10-20) | 59±3 | comparison only | ||
| 307 |  ToolACE-2-Llama-3.1-8B | 59±4 | local | ||
| 308 | CoALM-405B | 59±7 | local | ||
| 309 |  Gemma 2 9B IT DPO (Princeton) | 59±1 | local | ||
| 310 |  Ling-flash-2.0 | 59±2 | local | ||
| more from UIUC ConvAI & Oumi (CoALM) | |||||
| 606 | CoALM-70B | 49±4 | local | ||
| 627 | CoALM-8B | 48±4 | local | ||
$ what it's good at
agent tasks▁▁▃▆█▆▅▄▄▃▂▁ #204 of 334 * · 81% of its score · 1 test
Beats o1 (2024-12-17), o3-mini, DeepSeek R1 0528 here, though they rank higher overall. Trails SmolLM3 3B, Jamba 1.5 Mini, Mellum2 12B-A2.5B Instruct here, though they rank lower overall.
$ where to run it
Other ways to run it
- Run it as published · needs ~811.71 GB · from huggingface.co/uiuc-convai (official)
uv run --with torch --with 'transformers<5' --with accelerate python -c "from transformers import pipeline; print(pipeline('text-generation', model='uiuc-convai/CoALM-405B', torch_dtype='auto', device_map='auto')('Once upon a time', max_new_tokens=40)[0]['generated_text'])"
Official weights and documentation: huggingface.co/uiuc-convai ↗
$ how the score was worked out · Berkeley Function Calling Lea… 81%
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 |
|---|---|---|
 Berkeley Function Calling Leaderboard result archive | 1 | 81% |
| prior · pull toward the middle | 19% | |
| weighted average | = 59.1 |
$ 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