gpt-oss-120b
OpenAI- Arena score
- 1366
- Among open models
- #60of 148
- Among all on Arena
- #197
- Votes
- 30,115
Measured on gpt-oss-120b, the served version, not a local build. LMArena, CC BY 4.0, published Nov 11, 2026.
Score by category
CategoryScoreOpen rankArena rankVotes
- Math1389#56#1791.8K
- Spanish1388#51#163850
- Medicine & health1388#56#1761.6K
- Software & IT1387#62#19110.2K
- Applied math1387#56#1811.5K
- Chinese1385#62#1942.3K
- Science1382#61#1904.8K
- Coding1380#65#1986.4K
- English1375#66#20613K
- French1369#53#177459
- Hard prompts, English1368#70#2136.5K
- Hard prompts1363#62#20114.6K
- Business & finance1359#59#1875.3K
- Expert1355#64#1991.4K
- Legal & government1355#65#2051.9K
- German1354#52#171666
- Polish1353#47#1711.7K
- Non-English1350#59#19517.1K
- Russian1343#60#1992.1K
- Multi-turn1340#70#2184.9K
- Japanese1331#39#148574
- Excluding ties1330#60#19721.4K
- Long prompts1319#75#2346.5K
- Instructions1318#70#2207.8K
- Writing & literature1305#71#2286.6K
- Media & sports1292#75#2305.5K
- Korean1282#53#187680
- Creative writing1274#78#2473.9K
About as strong, smaller
Compare
gemma-4-26B-A4B-itGoogle143425.8B

gemma-3-27b-itGoogle135827.4B

Qwen3.8-27BQwen143727.8B

Qwen3.5-27BQwen140927.8B
Built on it
Ways to run it
- GGUF · llama.cpp, Ollama, LM Studioggml-org/gpt-oss-120b-GGUF49 builds
- MLX · Apple silicon: mlx-lm, LM Studiolmstudio-community/gpt-oss-120b-MLX-8bit41 builds
- GPTQ · vLLM, SGLangpositron-ai/openai_gpt-oss-120b-ingest-best-gptq1 build
Most downloaded
- gpt-oss-120b-GGUFggml-org · Quantization378K downloads
- gpt-oss-120b-Eagle3-short-contextnvidia · Fine-tune208.9K downloads
- gpt-oss-120b-GGUFunsloth · Quantization98.4K downloads
- gpt-oss-120b-GGUFlmstudio-community · Quantization44.3K downloads
- GPTOSS-120B-Uncensored-HauhauCS-AggressiveHauhauCS · Quantization42.7K downloads
BenchClub runs
Not runnable yet
Runs open once a gpt-oss-120b build is registered.
More from OpenAI
See more
gpt-oss-20b5.6M20.9B
Other 80–300B models
See more
DeepSeek-V4-FlashDeepSeek1.2M291B

MiniMax-M2.5MiniMax1M229B
Inkling-SmallThinking Machines1M266B
Hy3Tencent987.2K299B