The LLM benchmark aggregator.

GPT OSS 120B

OpenAI · 117B · open weights

Strongest in Secure code (#7 of 274), weakest in Toxicity avoidance (#236 of 272). Above par in 10 of 26 scopes. Among the models it meets almost everywhere, it finishes behind Claude Opus 5.5 and Claude Fable 5 and ahead of Gemma 4 31B and Claude Haiku 4.5.

§ 1 · Profile

What it is good at

Bars run from 50 — the average of the models each source lists — so right of the line is above par. The middle column is the gap to whoever leads that scope.

Safety54.2−7.3#69/3373/3

Secure code65.9−0.6#7/2741/1

Harm refusal57.7−3.9#25/3002/2

Fairness61.9−8.1#32/3002/2

Safe-prompt compliance46.8−13.6#63/821/1

Factual grounding39.3−31.3#87/1011/1

Jailbreak resistance42.8−22.1#211/2721/1

Toxicity avoidance43.4−15.6#236/2721/1

Reasoning48.6−18.3#96/1782/4

Mathematics57−8.7#28/1401/2

Science51.7−11.8#73/1221/1

Reasoning37.4−31.7#127/1401/3

Knowledge47.9−15.6#99/1381/2

Academic knowledge47.4−16.5#91/1231/1

Professional48.3−13.3#100/1681/1

Medical54.2−12.6#49/1401/1

Finance51−11.9#83/1511/1

Legal41.7−25.4#125/1511/1

Core abilities47.3−20.1#125/2042/3

General intelligence46.5−23.6#126/2042/3

Agents48.6−19.5#133/2681/5

Knowledge work47.9−25.7#91/1781/2

Coding41.9−27.2#142/1652/5

Agentic coding40.7−28#140/1572/4

Human preference52.4−15.1#163/3421/1

§ 2 · Head to head

What it beats, and what beats it

The same models turn up scope after scope. Counted once: where both were placed, who finished higher. Bars run right for GPT OSS 120B, left for the other.

GPT OSS 20Bwon 21 · lost 1

Gemini 1.5 Flashwon 15 · lost 2

Mistral Medium 3.5won 15 · lost 2

GPT-4o Miniwon 21 · lost 3

GPT-4.1 Nanowon 19 · lost 6

Llama 4 Scout Instructwon 18 · lost 6

Gemini 1.5 Prowon 14 · lost 5

Command R Pluswon 13 · lost 5

Mistral Largewon 13 · lost 5

Mistral Smallwon 14 · lost 6

Mistral Large 3won 14 · lost 6

Command Awon 16 · lost 7

Gemini 2.0 Flashwon 15 · lost 8

Gemini 2.5 Flash Litewon 12 · lost 7

Ling 3.0 Flashwon 11 · lost 7

Claude 3.5 Haikuwon 12 · lost 8

GPT-4owon 13 · lost 9

Grok 4.3won 14 · lost 10

GPT-5 Nanowon 13 · lost 10

O3 Miniwon 13 · lost 10

Grok 4.1 Fastwon 12 · lost 10

Llama 4 Maverick Instructwon 14 · lost 12

Claude Haiku 4.5won 14 · lost 12

Gemma 4 31Bwon 11 · lost 10

GPT-5.4 Nanowon 10 · lost 10

Kimi K2won 12 · lost 13

DeepSeek R1won 12 · lost 13

Grok 3won 10 · lost 11

Gemini 3.1 Flash Litewon 9 · lost 10

DeepSeek V3won 12 · lost 14

GPT-4.1 Miniwon 11 · lost 14

GPT-5.4 Miniwon 8 · lost 11

Kimi K2.6won 10 · lost 15

Mimo V2.5 Prowon 9 · lost 14

Grok 4 Fastwon 7 · lost 11

Qwen3 235B A22B Instructwon 9 · lost 15

MiniMax M3won 9 · lost 15

GLM-4.5won 7 · lost 12

O4 Miniwon 9 · lost 16

Gemini 2.5 Flashwon 9 · lost 16

Claude 3.5 Sonnetwon 7 · lost 13

Qwen3.6 Pluswon 6 · lost 12

Gemini 3.5 Flash Litewon 6 · lost 12

Grok 4won 8 · lost 16

DeepSeek V3.2won 8 · lost 17

Claude 3.7 Sonnetwon 7 · lost 15

Gemini 2.5 Prowon 8 · lost 18

GLM-4.6won 7 · lost 17

DeepSeek V4 Flashwon 7 · lost 17

GPT-5.6 Lunawon 7 · lost 17

GLM-5.3 Flashwon 7 · lost 17

Qwen3.7 Maxwon 7 · lost 17

GLM-5.2won 7 · lost 17

GPT-6 Lunawon 6 · lost 15

GPT-4.1won 7 · lost 18

Kimi K2.5won 7 · lost 18

GLM-5.1won 6 · lost 16

Claude Sonnet 4won 7 · lost 19

Mimo V2.6 Prowon 5 · lost 14

Kimi K2 Thinkingwon 5 · lost 15

Claude Opus 4won 6 · lost 18

Inklingwon 6 · lost 18

Gemini 3.5 Flashwon 6 · lost 18

Qwen3.8 Maxwon 6 · lost 18

Gemini 3.6 Flashwon 6 · lost 18

DeepSeek V4 Prowon 6 · lost 19

O1won 5 · lost 16

MiniMax M2.7won 5 · lost 16

Claude Sonnet 4.6won 5 · lost 18

Gemini 3.8 Flashwon 5 · lost 19

Gemini 3.1 Prowon 4 · lost 16

Claude Opus 4.6won 4 · lost 17

Inkling Smallwon 3 · lost 13

GLM-4.7won 3 · lost 13

GPT-6 Solwon 4 · lost 18

DeepSeek V4.1 Flashwon 3 · lost 14

Qwen3.8 27Bwon 3 · lost 15

GPT-5.6 Terrawon 4 · lost 20

GPT-5.4won 4 · lost 21

GPT-5.6 Solwon 4 · lost 21

GPT-5 Miniwon 4 · lost 22

Grok 4.7won 3 · lost 18

Gemini 3 Prowon 3 · lost 19

MiniMax M2.5won 2 · lost 14

O3won 3 · lost 21

Grok 4.5won 3 · lost 21

GPT-5.5won 3 · lost 22

Claude Sonnet 4.5won 3 · lost 23

GLM-5won 2 · lost 16

Claude Opus 4.5won 2 · lost 17

Grok 4.6won 2 · lost 22

GPT-5.1won 2 · lost 22

Claude Sonnet 5won 2 · lost 22

Muse Spark 1.1won 2 · lost 22

Claude Opus 4.7won 2 · lost 23

GPT-5won 2 · lost 24

Gemini 3 Flashwon 1 · lost 15

GPT-6 Astrawon 1 · lost 16

Muse Spark 1.2won 1 · lost 16

GLM-5.3won 1 · lost 17

Gemini 3.7 Flashwon 1 · lost 17

Kimi K3won 1 · lost 17

GPT-5.2won 1 · lost 23

Claude Opus 4.8won 1 · lost 23

Claude Opus 5won 1 · lost 23

Claude Fable 5.1won 0 · lost 18

Claude Fable 5won 0 · lost 18

Claude Opus 5.5won 0 · lost 21

§ 3 · Sources

Where the numbers come from

10 publications, 27 figures. Every one links to the page it was read from.

LMArena Text ↗1 measureread 2026-09-26

LMArena Text 1352

Artificial Analysis ↗1 measureread 2026-09-26

Artificial Analysis Intelligence Index 12

GDPval-AA ↗1 measureread 2026-09-26

GDPval-AA 596

Aider polyglot ↗1 measureread 2026-09-26

Aider polyglot 41.8%

Kagi LLM Benchmark ↗1 measureread 2026-09-26

Kagi LLM Benchmark 58.6%

SimpleBench ↗1 measureread 2026-09-26

SimpleBench 22.1%

Vals.ai ↗9 measuresread 2026-09-26

Vals · CaseLaw 48.77%Vals · LegalBench 75.94%Vals · CorpFin 58.24%Vals · TaxEval 71.59%Vals · MedQA 91.36%Vals · SWE-bench Verified 33.6% (Mini-SWE-agent)Vals · GPQA Diamond 78.54%Vals · MMLU Pro 79.17%Vals · AIME 92.6%

HELM Safety ↗5 measuresread 2026-09-26

HELM Safety · HarmBench 100%HELM Safety · SimpleSafetyTests 100%HELM Safety · Anthropic Red Team 99.5%HELM Safety · BBQ 98.5%HELM Safety · XSTest 92.7%

Enkrypt AI Safety Leaderboard ↗6 measuresread 2026-09-26

Enkrypt · Jailbreak risk 18.9%Enkrypt · Harmful content risk 14.4%Enkrypt · CBRN risk 13.8%Enkrypt · Toxicity risk 10.9%Enkrypt · Bias risk 60.2%Enkrypt · Insecure code risk 1.3%

Vectara Hallucination Leaderboard ↗1 measureread 2026-09-26

Vectara · Factual consistency 85.8%

Badge

[![PublicAI Index](https://publicai.io/model-index/badge?model=gpt-oss-120b)](https://publicai.io/model-index/m/gpt-oss-120b)