whatmodel.sh / OPT-30B

OPT-30B

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

OPT-30B is an open-weights language model from Meta, released May 2022. Its published size is 30B parameters and a 2k-token context window. Licence: opt-175b-license-noncommercial.

Catalogue identity notes

Meta's Open Pre-trained Transformer base model as released on Hugging Face (weights under Meta's non-commercial OPT licence).

28.7±1.4#653 of 719 LLMsfrom 7 benchmark results across 2 sources

$ how it compares

The LLMs ranked either side of OPT-30B, out of 828 with a score, then more from Meta.

#modelscorewhereRAMreleased
750StableLM-Base-Alpha 7B v2 Stability AI29±1local~30GB2023-08
751Granite 3.1 1B-A400M Instruct IBM29±2local~2.3GB2024-12
752OPT-30B Meta29±1local~64GB2022-05
753NexusRaven V2 13B Nexusflow29±3local~9.3GB2023-12
754OLMo 7B Allen Institute for AI28±1local~30GB2024-02
more from Meta
24Muse Spark 1.3 Meta77±1api $2.00/M2026-09
40Muse Spark 1.2 Meta75±1api $2.00/M2026-08
43Muse Spark 1.1 Meta75±1api $2.00/M2026-07

see all LLMs · find the ones that fit your machine

$ what it's good at

other tests7 not in a task · 99% of its score
testresultvs the others in its rundifficultycountswhy
ARC Challenge34.6% ✎█··▁·▁▁·▁▂▁▃ beats 18/19medium · avg 72%▇▇▇▇▇▇▇▇▇▇▇▇ 29%sharp
WinoGrande68.2% ✎▄····▄▆▂▄▆▆█ beats 18/19easy · avg 89%▇▇▇▇▇▇▇▇▇ 22%at its level
ARC Easy70.0% ✎▃·▅▃▃▅·██▃██ beats 17/19very easy · avg 90%▇▇▇▇▇▇▇▇ 19%at its level
PIQA77.6% ✎▂·▄▄·▄▂▆▂▄█▄ beats 18/19easy · avg 88%▇▇▇▇ 11%blunt · at its level
SciQ91.1% ✎▂··▂▃▂▂▆▃▃█▂ beats 18/19very easy · avg 98%▇▇▇ 8.7%typical
LAMBADA (OpenAI version), last-word accuracy71.5% ✎▃··▅▃▃▅▅▅███ beats 18/19very easy · avg 89%▇▇▇ 7.7%blunt · at its level
Open PT-LLM Leaderboard average (9 Portuguese tasks)39.5% ✎▃▁▁▁▁▃▁▁▂█▅▁ beats 48/276medium · avg 73%▇ 2.5%noisy

$ where to run it

Other ways to run it

  1. Run it as published · needs ~59.95 GB · from huggingface.co/facebook
    uv run --with torch --with 'transformers<5' --with accelerate python -c "from transformers import pipeline; print(pipeline('text-generation', model='facebook/opt-30b', torch_dtype='auto', device_map='auto')('Once upon a time', max_new_tokens=40)[0]['generated_text'])"

Weights and documentation: huggingface.co/facebook ↗

$ how the score was worked out · Pythia: A Suite for Analyzing… 97%, Open Portuguese LLM Leaderboa… 3%

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

sourceresultsshare of the scorelicence
Pythia: A Suite for Analyzing Large Language Models Across Training and Scaling — authors' own runs6▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 97%CC BY-SA 4.0
Open Portuguese LLM Leaderboard (archived)1▇ 3%Apache-2.0
prior · pull toward the middle▇ 1%
weighted average= 28.7

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

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