The LLM benchmark aggregator.

Claude 3 Opus

Anthropic

Strongest in Harm refusal (#7 of 300), weakest in Human preference (#201 of 342). Above par in 6 of 11 scopes. Among the models it meets almost everywhere, it finishes behind Claude 3.5 Sonnet and Claude Opus 5.5 and ahead of Gemini 2.5 Pro and Claude Opus 4.

§ 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.

Safety59.6−1.9#10/3372/3

Harm refusal60.1−1.5#7/3002/2

Jailbreak resistance63.4−1.5#7/2721/1

Toxicity avoidance58.8−0.2#9/2721/1

Fairness64.7−5.3#25/3002/2

Safe-prompt compliance46.4−14#67/821/1

Secure code59.1−7.4#74/2741/1

Reasoning42.5−24.4#142/1781/4

Reasoning38−31.1#122/1401/3

Human preference49.5−18#201/3421/1

§ 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 Claude 3 Opus, left for the other.

OLMoE 1B-7B Instructwon 7 · lost 0

Mistral Instruct v0.1 (7B)won 7 · lost 0

OLMo 2 7B Instructwon 7 · lost 0

IBM Granite 4.0 Microwon 7 · lost 0

Jamba 1.6 Miniwon 7 · lost 0

Jamba 1.6 Largewon 7 · lost 0

Llama 3.3 Instruct Turbo (70B)won 7 · lost 0

Gemma 2B Itwon 8 · lost 0

Llama 3.2 1B Instructwon 8 · lost 0

Smollm2 1.7B Instructwon 8 · lost 0

Phi 3 Mini 128k Instructwon 8 · lost 0

Zephyr 7B Betawon 8 · lost 0

Gemma 7B Itwon 8 · lost 0

Phi 3 Mini 4k Instructwon 8 · lost 0

Mistral 7B Instruct V0.2won 8 · lost 0

Granite 3.0 2B Instructwon 8 · lost 0

Nous Hermes 2 Mixtral 8x7b Dpowon 8 · lost 0

Llama 3.2 3B Instructwon 8 · lost 0

Starling Lm 7B Betawon 8 · lost 0

Granite 3.1 2B Instructwon 8 · lost 0

Granite 3.0 8B Instructwon 8 · lost 0

Qwen1.5 14B Chatwon 8 · lost 0

Phi 3 Medium 4k Instructwon 8 · lost 0

Granite 3.1 8B Instructwon 8 · lost 0

Llama 3.1 8B Instructwon 8 · lost 0

Llama 3.1 Tulu 3 8Bwon 8 · lost 0

C4ai Aya Expanse 8Bwon 8 · lost 0

Nova Microwon 8 · lost 0

Phi 4won 8 · lost 0

Nova Litewon 8 · lost 0

Gemma 2 9B Itwon 8 · lost 0

Mistral Small 24B Instructwon 8 · lost 0

GPT-4won 8 · lost 0

Jamba 1.5 Largewon 8 · lost 0

Nova Prowon 8 · lost 0

Gemma 3 4B Itwon 8 · lost 0

Qwen Maxwon 8 · lost 0

DeepSeek LLM 67B Chatwon 9 · lost 0

DBRX Instructwon 9 · lost 0

Claude 3 Haikuwon 9 · lost 0

Gemini 1.5 Flashwon 10 · lost 0

GPT-3.5 Turbowon 10 · lost 1

Command R Pluswon 10 · lost 1

GPT-4o Miniwon 10 · lost 1

Llama 3 8B Instructwon 8 · lost 1

Qwen2 72B Instructwon 8 · lost 1

Claude 3 Sonnetwon 8 · lost 1

Mistral Largewon 8 · lost 1

Phi 3 Small 8k Instructwon 7 · lost 1

Gemma 2 2B Itwon 7 · lost 1

Jamba 1.5 Miniwon 7 · lost 1

C4ai Aya Expanse 32Bwon 7 · lost 1

Qwen2.5 72B Instructwon 7 · lost 1

Mistral Instruct v0.3 (7B)won 6 · lost 1

Llama 3.1 Instruct Turbo (70B)won 6 · lost 1

Llama 3.1 Instruct Turbo (405B)won 6 · lost 1

Qwen 2.5 Instruct Turbo (7B)won 6 · lost 1

Ling 3.0 Flashwon 6 · lost 1

Magistral Mediumwon 8 · lost 2

GPT OSS 20Bwon 8 · lost 2

Llama 3 70B Instructwon 7 · lost 2

Command Awon 7 · lost 2

Mistral Smallwon 7 · lost 2

Grok Build 0.1won 6 · lost 2

Nemotron 3 Ultra 550B A55Bwon 6 · lost 2

Magistral Smallwon 6 · lost 2

Kimi K2 Thinkingwon 6 · lost 2

Gemma 2 27B Itwon 6 · lost 2

QwQ 32Bwon 6 · lost 2

Gemma 3 12B Itwon 6 · lost 2

Gemma 3 27B Itwon 6 · lost 2

Qwen3 30B A3B Instructwon 6 · lost 2

Gemini 1.5 Prowon 8 · lost 3

Gemini 2.0 Flashwon 8 · lost 3

Llama 3.1 Instruct Turbo (8B)won 5 · lost 2

Qwen 2.5 Instruct Turbo (72B)won 5 · lost 2

Gemma 4 E4Bwon 5 · lost 2

Gemma 4 12Bwon 5 · lost 2

MiniMax M3won 7 · lost 3

Grok 4.3won 7 · lost 3

Gemini 2.0 Flash Litewon 6 · lost 3

GLM-4.5won 6 · lost 3

GLM-4.6won 6 · lost 3

Mimo V2.6 Flashwon 6 · lost 3

Mimo V2.5 Prowon 6 · lost 3

Mimo V2.6 Prowon 6 · lost 3

Llama 4 Scout Instructwon 7 · lost 4

Llama 4 Maverick Instructwon 7 · lost 4

GPT-4owon 7 · lost 4

O1 Miniwon 7 · lost 4

O3 Miniwon 7 · lost 4

DeepSeek V3won 7 · lost 4

Kimi K2.7 Codewon 5 · lost 3

NVIDIA Nemotron 3 Super 120B A12Bwon 5 · lost 3

Qwen3.5 35B A3Bwon 5 · lost 3

Qwen3.5 27Bwon 5 · lost 3

Qwen3 Vl 235B A22B Instructwon 5 · lost 3

Qwen3.5 122B A10Bwon 5 · lost 3

Gemma 4 26B A4Bwon 5 · lost 3

Grok 3won 6 · lost 4

DeepSeek V3.2won 6 · lost 4

Grok 4.1 Fastwon 6 · lost 4

DeepSeek V4 Flashwon 6 · lost 4

GPT-6 Solwon 6 · lost 4

DeepSeek V4 Prowon 6 · lost 4

Gemini 3.6 Flashwon 6 · lost 4

Claude 3.5 Haikuwon 5 · lost 4

Gemma 4 31Bwon 5 · lost 4

GPT-4.1 Nanowon 6 · lost 5

GPT OSS 120Bwon 6 · lost 5

Gemini 2.5 Flashwon 6 · lost 5

Grok 4won 6 · lost 5

Claude Haiku 4.5won 6 · lost 5

GPT-4.1won 6 · lost 5

DeepSeek R1won 6 · lost 5

Claude Opus 4won 6 · lost 5

Gemini 2.5 Prowon 6 · lost 5

Qwen3.5 397B A17Bwon 4 · lost 4

Claude Sonnet 4.6won 4 · lost 4

Claude Opus 4.6won 4 · lost 4

GPT-5.2won 5 · lost 5

GPT-6 Lunawon 5 · lost 5

Kimi K2.5won 5 · lost 5

GPT-5.6 Lunawon 5 · lost 5

Kimi K2.6won 5 · lost 5

GLM-5.1won 5 · lost 5

Grok 4.5won 5 · lost 5

GPT-5.6 Terrawon 5 · lost 5

GLM-5.3 Flashwon 5 · lost 5

Qwen3.7 Maxwon 5 · lost 5

GPT-5.4won 5 · lost 5

GLM-5.2won 5 · lost 5

Gemini 3.5 Flashwon 5 · lost 5

Qwen3.8 Maxwon 5 · lost 5

GPT-5.5won 5 · lost 5

GPT-5.6 Solwon 5 · lost 5

Gemini 3.8 Flashwon 5 · lost 5

Claude Opus 4.7won 5 · lost 5

GPT-4 Turbowon 5 · lost 6

GPT-4.1 Miniwon 5 · lost 6

Claude 3.7 Sonnetwon 5 · lost 6

GPT-5 Miniwon 5 · lost 6

O4 Miniwon 5 · lost 6

Claude Sonnet 4won 5 · lost 6

Kimi K2won 5 · lost 6

Qwen3 235B A22B Instructwon 5 · lost 6

O3won 5 · lost 6

GPT-5.1won 5 · lost 6

Claude Sonnet 4.5won 5 · lost 6

Gemini 2.5 Flash Litewon 3 · lost 4

MiniMax M2won 4 · lost 6

Grok 4.6won 4 · lost 6

Claude Opus 4.8won 4 · lost 6

Claude Opus 5won 4 · lost 6

GPT-4.5won 3 · lost 5

Hy3won 3 · lost 5

Gemini 3 Prowon 3 · lost 5

GPT-5 Nanowon 4 · lost 7

O1won 4 · lost 7

GPT-5won 4 · lost 7

MiniMax M2.7won 3 · lost 6

Grok 4.7won 3 · lost 7

Inklingwon 3 · lost 7

Claude Sonnet 5won 3 · lost 7

Muse Spark 1.1won 3 · lost 7

Claude Opus 5.5won 3 · lost 7

Claude 3.5 Sonnetwon 2 · lost 9

§ 3 · Sources

Where the numbers come from

4 publications, 13 figures. Every one links to the page it was read from.

LMArena Text ↗1 measureread 2026-09-26

LMArena Text 1322

SimpleBench ↗1 measureread 2026-09-26

SimpleBench 23.5%

HELM Safety ↗5 measuresread 2026-09-26

HELM Safety · HarmBench 97.4%HELM Safety · SimpleSafetyTests 100%HELM Safety · Anthropic Red Team 99.8%HELM Safety · BBQ 94%HELM Safety · XSTest 92.5%

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

Enkrypt · Jailbreak risk 1.5%Enkrypt · Harmful content risk 7.8%Enkrypt · CBRN risk 4.3%Enkrypt · Toxicity risk 0.1%Enkrypt · Bias risk 25.6%Enkrypt · Insecure code risk 15.1%

Badge

[![PublicAI Index](https://publicai.io/model-index/badge?model=claude-3-opus)](https://publicai.io/model-index/m/claude-3-opus)