whatmodel.sh / StableLM-Base-Alpha 3B v2
StableLM-Base-Alpha 3B v2
↗ share
+ contribute
StableLM-Base-Alpha 3B v2 is an open-weights language model from Stability AI, released August 2023. Its published size is 2.8B parameters and a 4k-token context window. Licence: cc-by-sa-4.0.
Catalogue identity notes
Maker repo stabilityai/stablelm-base-alpha-3b-v2 @997d959: licence tag cc-by-sa-4.0. 2,795,934,720 parameters from the Hugging Face safetensors metadata (dense, so active = total). Released 2023-08: month the maker's Hugging Face repo was created (createdAt 2023-08-04). Pretrained base checkpoint (not an instruct/chat version).
$ how it compares
The LLMs ranked either side of StableLM-Base-Alpha 3B v2, out of 828 with a score, then more from Stability AI.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 783 |  RWKV-4-Raven 14B | 24±2 | comparison only | ||
| 784 |  Falcon3 1B Instruct | 24±4 | local | ||
| 785 | StableLM-Base-Alpha 3B v2 | 24±1 | local | ||
| 786 |  Qwen1.5 1.8B Chat | 24±2 | local | ||
| 787 |  Pythia 6.9B | 23±1 | local | ||
| more from Stability AI | |||||
| 718 | Stable Code 3B | 33±1 | local | ||
| 746 | StableLM-3B-4E1T | 29±1 | local | ||
| 750 | StableLM-Base-Alpha 7B v2 | 29±1 | local | ||
$ what it's good at
other tests12 not in a task · 100% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| ARC Challenge lab | 35.1% ✎ | ▁·▁··▁▁▇▂█▆▃ beats 7/26 | 37% |
| WinoGrande lab | 62.1% ✎ | ▁····▁··▁▃█▄ beats 6/26 | 12% |
| BoolQ lab | 64.6% ✎ | ▂▃▂▅▆▂▅▆█▂▂▂ beats 7/26 | 11% |
| ARC Easy lab | 63.3% ✎ | ▁▁·▁▁▄▅▄█▅▁▃ beats 9/26 | 8.5% |
| SciQ lab | 92.1% ✎ | ▁·▁·▁·▃▅▃█▄▁ beats 11/26 | 8.2% |
| HellaSwag lab | 68.6% ✎ | ▃···▁▇▅█▇▃▁▁ beats 11/26 | 7.2% |
| PIQA lab | 76.0% ✎ | ▁·▁··▁▁▃█▅█▂ beats 9/26 | 5.3% |
| LAMBADA (OpenAI version), last-word accuracy lab | 70.3% ✎ | ▁··▂▁▁▁▂▆█▅▃ beats 14/26 | 3.5% |
4 more tests
| test | result | vs the others in its run | counts |
|---|---|---|---|
| OpenBookQA lab | 26.4% ✎ | █·▁▁▃·▁▃▅▇▃▄ beats 8/26 | 2.7% |
| Social IQA lab | 42.5% ✎ | ▁···▁▃█▁▁▁·▁ beats 16/26 | 2.5% |
| Open PT-LLM Leaderboard average (9 Portuguese tasks) | 30.6% ✎ | ▃▁▁▁▁▁▁▁▂█▅▁ beats 32/276 | 1.1% |
| TruthfulQA MC2 lab | 35.9% ✎ | █·█·█▄···▄·█ beats 5/9 | 0.6% |
$ where to run it
Other ways to run it
- Run it as published · needs ~11.18 GB · from huggingface.co/stabilityai (official)
uv run --with torch --with 'transformers<5' --with accelerate python -c "from transformers import pipeline; print(pipeline('text-generation', model='stabilityai/stablelm-base-alpha-3b-v2', torch_dtype='auto', device_map='auto', trust_remote_code=True)('Once upon a time', max_new_tokens=40)[0]['generated_text'])"
Official weights and documentation: huggingface.co/stabilityai ↗
$ how the score was worked out · Stability AI — original Stabl… 98%, Open Portuguese LLM Leaderboa… 1%
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 |
|---|---|---|
 Stability AI — original StableLM zero-shot evaluation outputs · lab-reported | 11 | 98% |
 Open Portuguese LLM Leaderboard (archived) | 1 | 1% |
| prior · pull toward the middle | 0% | |
| weighted average | = 23.9 |
$ 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