Request a pilot
‹ PublicAI Index

The LLM benchmark aggregator.

Claude Opus 5.5

Anthropic

Strongest in Human preference (#1 of 342), weakest in Toxicity avoidance (#119 of 272). Above par in 26 of 27 scopes. Among the models it meets almost everywhere, it finishes behind none and ahead of Claude Fable 5 and Claude Fable 5.1.

§ 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.

Human preference67.5#1/342
Human preference67.5#1/342
Agents68.1#1/268
Knowledge work73.6#1/178
Reasoning66.9#1/178
Reasoning69.1#1/140
Mathematics65.7#1/140
Coding69.1#1/165
Agentic coding68.7#1/157
Code generation70#1/77
Safety60.6#2/337
Secure code66.3#2/274
Fairness70#3/300
Harm refusal58.2#23/300
Jailbreak resistance59.8#49/272
Toxicity avoidance54.6#119/272
Core abilities62.2#6/204
General intelligence70.1#1/204
Data analysis61.6#6/57
Language63.2#8/57
Instruction following46.1#35/57
Professional59.7#7/168
Biology research68.5#1/16
IT operations62.3#2/11
Medical61.1#5/140
Finance58.1#18/151
Legal55.5#38/151

§ 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 Opus 5.5, left for the other.

Grok 3won 17 · lost 0
Gemini 3 Prowon 17 · lost 0
GPT OSS 20Bwon 18 · lost 0
Grok 4.1 Fastwon 18 · lost 0
Grok Build 0.1won 19 · lost 0
Command Awon 19 · lost 0
Gemini 2.0 Flashwon 19 · lost 0
Qwen3.6 Pluswon 19 · lost 0
GPT-5.4 Miniwon 19 · lost 0
Gemma 4 31Bwon 19 · lost 0
GLM-5.1won 19 · lost 0
Kimi K2won 20 · lost 0
Qwen3 235B A22B Instructwon 20 · lost 0
GLM-4.6won 20 · lost 0
GPT-4.1 Nanowon 21 · lost 0
GPT OSS 120Bwon 21 · lost 0
GPT-4.1 Miniwon 21 · lost 0
O4 Miniwon 21 · lost 0
Gemini 2.5 Flashwon 21 · lost 0
DeepSeek V3.2won 21 · lost 0
Kimi K2.5won 21 · lost 0
DeepSeek V3won 22 · lost 0
Claude Sonnet 4.6won 23 · lost 0
Grok 4.3won 25 · lost 0
Kimi K2.6won 25 · lost 0
DeepSeek V4 Flashwon 26 · lost 0
GPT-6 Lunawon 26 · lost 0
GPT-5.6 Lunawon 26 · lost 0
GLM-5.2won 25 · lost 1
Gemini 3.6 Flashwon 25 · lost 1
MiniMax M3won 24 · lost 1
GPT-5.6 Terrawon 24 · lost 1
Qwen3.7 Maxwon 24 · lost 1
DeepSeek R1won 21 · lost 1
Gemini 2.5 Prowon 21 · lost 1
GPT-4o Miniwon 20 · lost 1
O3 Miniwon 19 · lost 1
GPT-5.4 Nanowon 19 · lost 1
Grok 4won 19 · lost 1
GPT-4.1won 19 · lost 1
Mimo V2.5 Prowon 19 · lost 1
Gemini 3.1 Prowon 19 · lost 1
Kimi K2.7 Codewon 18 · lost 1
Llama 4 Scout Instructwon 18 · lost 1
Gemini 3.5 Flash Litewon 18 · lost 1
Claude Opus 4.5won 18 · lost 1
GPT-4owon 17 · lost 1
Grok 4.6won 24 · lost 2
GPT-6 Solwon 24 · lost 2
Grok 4.5won 24 · lost 2
DeepSeek V4 Prowon 23 · lost 2
GPT-5.4won 23 · lost 2
Qwen3.8 Maxwon 23 · lost 2
Claude Sonnet 4won 20 · lost 2
Claude Haiku 4.5won 19 · lost 2
O3won 19 · lost 2
Claude Sonnet 4.5won 19 · lost 2
DeepSeek V4.1 Flashwon 19 · lost 2
Claude Opus 4won 18 · lost 2
Qwen3.8 27Bwon 17 · lost 2
Mimo V2.6 Prowon 17 · lost 2
GLM-5.3won 17 · lost 2
GPT-5 Nanowon 16 · lost 2
MiniMax M2.7won 16 · lost 2
GPT-5.6 Solwon 24 · lost 3
Kimi K2 Thinkingwon 15 · lost 2
GLM-5.3 Flashwon 22 · lost 3
Gemini 3.5 Flashwon 22 · lost 3
GPT-5.2won 21 · lost 3
GPT-5 Miniwon 18 · lost 3
Claude Opus 4.6won 18 · lost 3
Gemini 3.7 Flashwon 17 · lost 3
Kimi K3won 17 · lost 3
Grok 4.7won 22 · lost 4
Gemini 3.8 Flashwon 22 · lost 4
GPT-5.1won 16 · lost 3
Inklingwon 21 · lost 4
Claude 3.7 Sonnetwon 15 · lost 3
Claude 3.5 Haikuwon 14 · lost 3
Claude 3.5 Sonnetwon 14 · lost 3
O1won 14 · lost 3
GPT-5won 17 · lost 4
Muse Spark 1.2won 16 · lost 4
Claude Sonnet 5won 20 · lost 5
Claude Opus 5won 21 · lost 6
GPT-6 Astrawon 17 · lost 5
GPT-5.5won 20 · lost 6
Claude Opus 4.8won 19 · lost 6
Muse Spark 1.1won 19 · lost 6
Claude Opus 4.7won 19 · lost 6
Muse Spark 1.3won 13 · lost 5
Claude Fable 5.1won 12 · lost 8
Claude Fable 5won 11 · lost 8

§ 3 · Sources

Where the numbers come from

9 publications, 30 figures. Every one links to the page it was read from.

LMArena Text ↗1 measureread 2026-09-26
LMArena Text 1509
Artificial Analysis ↗1 measureread 2026-09-26
Artificial Analysis Intelligence Index 58
GDPval-AA ↗1 measureread 2026-09-26
GDPval-AA 1846
AA-Briefcase ↗1 measureread 2026-09-26
AA-Briefcase 1822
ARC-AGI-2 ↗1 measureread 2026-09-26
ARC-AGI-2 91.7%
LiveBench ↗8 measuresread 2026-09-26
LiveBench 83.2LiveBench · Reasoning 92.2LiveBench · Coding 89.3LiveBench · Agentic Coding 71.7LiveBench · Mathematics 97.1LiveBench · Data Analysis 80.3LiveBench · Language 86.3LiveBench · Instruction Following 65.7
SimpleBench ↗1 measureread 2026-09-26
SimpleBench 88.4%
Vals.ai ↗10 measuresread 2026-09-26
Vals · Legal Research Bench 50.48%Vals · Harvey Legal Agent Benchmark 3.75%Vals · Finance Agent 58.59%Vals · MedCode 49.8%Vals · MedScribe 91.43%Vals · BioMysteryBench 79.26%Vals · SRE Bench 33.59%Vals · Vibe Code Bench 90.29% (OpenHands)Vals · Code Migration 66.65%Vals · ProofBench 100%
Enkrypt AI Safety Leaderboard ↗6 measuresread 2026-09-26
Enkrypt · Jailbreak risk 4.5%Enkrypt · Harmful content risk 1.7%Enkrypt · CBRN risk 10.2%Enkrypt · Toxicity risk 3.1%Enkrypt · Bias risk 25.6%Enkrypt · Insecure code risk 0.4%

Badge

PublicAI Index badge for Claude Opus 5.5[![PublicAI Index](https://publicai.io/model-index/badge?model=claude-opus-5-5)](https://publicai.io/model-index/m/claude-opus-5-5)