[whatmodel.sh](../../README.md) / StableLM-Base-Alpha 7B v2 # StableLM-Base-Alpha 7B v2 Stability AI · released 2023-08 · 6.89B params · cc-by-sa-4.0 **score 29.5** ±1.0 · #667 of 732 · from 12 benchmark results across 2 sources Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only. ## What it's good at _{● marks StableLM-Base-Alpha 7B v2 among every model on that task or test, worst to best.}

other tests (12)

test                                result  vs the others in its run   difficulty           counts  why
ARC Challenge (maker)                40.5%  ▁·▁··▁▁▇▂●▆▃ beats 14/26   easy · avg 82%          48%  sharp
WinoGrande (maker)                   68.8%  ▁····▁··▁▂█● beats 21/26   easy · avg 86%          11%  at its level
BoolQ (maker)                        70.3%  ▂▃▂▅▆▂●▆█▂▂▂ beats 13/26   easy · avg 88%         8.5%  at its level
HellaSwag (maker)                    74.3%  ▃···▁▇▅●▇▃▁▁ beats 17/26   very easy · avg 91%    6.2%  at its level
ARC Easy (maker)                     69.1%  ▁▁·▁▁▄▅▄●▅▁▃ beats 16/26   easy · avg 85%         5.8%  blunt · at its level
SciQ (maker)                         93.9%  ▁·▁·▁·▃▅▃●▄▁ beats 20/26   very easy · avg 97%    4.8%  blunt
PIQA (maker)                         78.5%  ▁·▁··▁▁▃█▅●▂ beats 18/26   easy · avg 87%         4.2%  blunt · at its level
OpenBookQA (maker)                   30.4%  █·▁▁▃·▁▃●▇▃▄ beats 15/26   hard · avg 49%         3.8%  too hard for it
LAMBADA (OpenAI version), last-wo…   74.2%  ▁··▂▁▁▁▂▆█●▃ beats 23/26   very easy · avg 86%    2.6%  blunt · at its level
Social IQA (maker)                   42.4%  ▁···▁▃●▁▁▁·▁ beats 15/26   hard · avg 56%         2.4%  blunt
Open PT-LLM Leaderboard average (…   44.6%  ▃▁▁▁▁▁●▁▂█▅▁ beats 58/276  medium · avg 72%       1.7%  noisy
TruthfulQA MC2 (maker)               36.5%  █·█·█●···▄·█ beats 6/9     medium · avg 55%       0.4%  blunt · at its level

## How the score was worked out Each result implies a score on its own, after adjusting for how hard that test is and who ran it. The model's 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 is added up by who ran the tests. [Method](../method.md)

source                                    results  share of the score
Stability AI (submitted / maker-reporte…       11   98% ▇▇▇▇▇▇▇▇▇▇▇▇
Open Portuguese LLM Leaderboard                 1    2% ▇
prior: pull toward the middle                        0% ▇
weighted average                                   = 29.5

_{Licences: Stability AI ([Apache-2.0](../../licenses/stablelm-original-eval-json-Apache-2.0.txt)) · Open Portuguese LLM Leaderboard ([Apache-2.0](../../licenses/hf-leaderboard-spaces-Apache-2.0.txt))} ## How to run it ### Other ways to run it 1. Run it as published (fp32, needs ~27.56 GB, from [huggingface.co/stabilityai](https://huggingface.co/stabilityai/stablelm-base-alpha-7b-v2/tree/6d176503c1496b3c96a2bceb76a7d37695582618)) ```sh uv run --with torch --with 'transformers<5' --with accelerate python -c "from transformers import pipeline; print(pipeline('text-generation', model='stabilityai/stablelm-base-alpha-7b-v2', torch_dtype='auto', device_map='auto', trust_remote_code=True)('Once upon a time', max_new_tokens=40)[0]['generated_text'])" ``` Model repository and documentation: [huggingface.co/stabilityai](https://huggingface.co/stabilityai/stablelm-base-alpha-7b-v2) [interactive version](https://whatmodel.sh/model/stablelm-base-alpha-7b-v2) · [JSON](https://whatmodel.sh/api?format=json&all=1)