Request a pilot
‹ PublicAI Index

The LLM benchmark aggregator.

Claude Sonnet 4.6

Anthropic

Strongest in Computer use (#8 of 37), weakest in Secure code (#154 of 274). Above par in 22 of 28 scopes. Among the models it meets almost everywhere, it finishes behind Claude Opus 5.5 and Claude Fable 5 and ahead of O3 and GPT-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.

Professional55.8#30/168
Finance57.7#22/151
Legal55.1#41/151
Medical54.6#45/140
Knowledge56.4#32/138
Academic knowledge57.6#29/123
Human preference63.9#35/342
Human preference63.9#35/342
Agents58.8#40/268
Computer use64.2#8/37
Knowledge work57.9#41/178
Reasoning53.7#51/178
Reasoning53.7#46/140
Science56.6#48/122
Mathematics52.4#70/140
Coding51.3#65/165
Code generation53.9#31/77
Agentic coding50.6#69/157
Core abilities47.6#120/204
Data analysis57.6#22/57
Language44#39/57
Instruction following41.7#43/57
General intelligence47.1#123/204
Safety52#131/337
Harm refusal56.5#42/300
Factual grounding48.4#63/101
Fairness47.9#145/300
Secure code50.3#154/274

§ 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 Sonnet 4.6, left for the other.

Mistral Medium 3.5won 17 · lost 0
Mistral Large 3won 20 · lost 0
Gemini 3.5 Flash Litewon 21 · lost 1
Llama 4 Scout Instructwon 20 · lost 1
Command Awon 20 · lost 1
Gemini 2.0 Flashwon 19 · lost 1
Mistral Smallwon 17 · lost 1
Grok 4.3won 24 · lost 2
GPT-5.4 Miniwon 21 · lost 2
GPT-4o Miniwon 20 · lost 2
GPT-4.1 Nanowon 20 · lost 2
Kimi K2.7 Codewon 15 · lost 2
Grok Build 0.1won 15 · lost 2
Qwen3.6 Pluswon 19 · lost 3
GPT OSS 20Bwon 16 · lost 3
Claude 3.5 Haikuwon 15 · lost 3
Gemini 2.5 Flash Litewon 15 · lost 3
GPT-5.4 Nanowon 20 · lost 4
Kimi K2.5won 20 · lost 4
Mimo V2.5 Prowon 17 · lost 4
GPT-4owon 16 · lost 4
Grok 4.1 Fastwon 16 · lost 4
DeepSeek V3won 19 · lost 5
Grok 3won 15 · lost 4
Gemini 3.1 Flash Litewon 15 · lost 4
GPT OSS 120Bwon 18 · lost 5
Claude Haiku 4.5won 18 · lost 5
Grok 4 Fastwon 14 · lost 4
GPT-4.1 Miniwon 17 · lost 5
Inklingwon 20 · lost 6
GLM-4.5won 13 · lost 4
Claude 3.7 Sonnetwon 15 · lost 5
Gemma 4 31Bwon 15 · lost 5
GLM-5.1won 15 · lost 5
Kimi K2.6won 21 · lost 7
DeepSeek R1won 17 · lost 6
GLM-4.6won 16 · lost 6
Claude 3.5 Sonnetwon 13 · lost 5
GLM-5won 13 · lost 5
MiniMax M3won 19 · lost 8
Gemini 2.5 Flashwon 16 · lost 7
DeepSeek V3.2won 16 · lost 7
MiniMax M2.7won 13 · lost 6
Kimi K2won 15 · lost 7
Gemini 2.5 Prowon 16 · lost 8
GPT-5 Miniwon 15 · lost 8
GPT-6 Lunawon 15 · lost 8
Claude Sonnet 4won 16 · lost 9
Claude Opus 4won 14 · lost 8
GLM-5.2won 16 · lost 10
O1won 11 · lost 7
GPT-5 Nanowon 12 · lost 8
Qwen3 235B A22B Instructwon 13 · lost 9
Qwen3.8 27Bwon 13 · lost 9
Claude Sonnet 4.5won 14 · lost 10
O3 Miniwon 12 · lost 9
O4 Miniwon 13 · lost 10
Kimi K2 Thinkingwon 10 · lost 8
GPT-4.1won 12 · lost 10
DeepSeek V4 Flashwon 14 · lost 12
GPT-5.6 Lunawon 14 · lost 12
Qwen3.7 Maxwon 14 · lost 12
GPT-5.1won 11 · lost 10
O3won 12 · lost 11
DeepSeek V4 Prowon 13 · lost 14
Claude Opus 4.5won 11 · lost 12
Grok 4won 10 · lost 11
Gemini 3.5 Flashwon 12 · lost 14
Gemini 3.6 Flashwon 12 · lost 14
GPT-5won 10 · lost 13
GPT-5.2won 11 · lost 15
GLM-5.3 Flashwon 11 · lost 15
Gemini 3.1 Prowon 10 · lost 14
Grok 4.7won 9 · lost 14
Grok 4.5won 10 · lost 16
Gemini 3 Prowon 8 · lost 13
GPT-5.6 Terrawon 9 · lost 17
GLM-5.3won 7 · lost 15
GPT-5.4won 8 · lost 19
Muse Spark 1.1won 7 · lost 20
GPT-6 Solwon 6 · lost 18
Claude Opus 4.6won 6 · lost 19
DeepSeek V4.1 Flashwon 5 · lost 16
Mimo V2.6 Prowon 4 · lost 13
Grok 4.6won 6 · lost 20
Claude Sonnet 5won 5 · lost 21
Gemini 3.8 Flashwon 5 · lost 21
Qwen3.8 Maxwon 4 · lost 22
GPT-5.6 Solwon 4 · lost 23
GPT-6 Astrawon 3 · lost 18
Muse Spark 1.2won 3 · lost 18
Gemini 3.7 Flashwon 3 · lost 19
GPT-5.5won 2 · lost 25
Claude Opus 5won 2 · lost 25
Kimi K3won 1 · lost 21
Claude Opus 4.8won 1 · lost 25
Claude Opus 4.7won 1 · lost 26
Muse Spark 1.3won 0 · lost 18
Claude Fable 5.1won 0 · lost 22
Claude Fable 5won 0 · lost 23
Claude Opus 5.5won 0 · lost 23

§ 3 · Sources

Where the numbers come from

9 publications, 33 figures. Every one links to the page it was read from.

LMArena Text ↗1 measureread 2026-09-26
LMArena Text 1472
GDPval-AA ↗1 measureread 2026-09-26
GDPval-AA 1220
AA-Briefcase ↗1 measureread 2026-09-26
AA-Briefcase 1063
ARC-AGI-2 ↗1 measureread 2026-09-26
ARC-AGI-2 58.3%
LiveBench ↗8 measuresread 2026-09-26
LiveBench 73LiveBench · Reasoning 84.8LiveBench · Coding 79.3LiveBench · Agentic Coding 42.6LiveBench · Mathematics 87LiveBench · Data Analysis 77.9LiveBench · Language 76.1LiveBench · Instruction Following 63.2
OSWorld ↗1 measureread 2026-09-26
OSWorld 72.1%
Vals.ai ↗15 measuresread 2026-09-26
Vals · Legal Research Bench 38.46%Vals · CaseLaw 63.99%Vals · LegalBench 82.12%Vals · Harvey Legal Agent Benchmark 5%Vals · Finance Agent 51.03%Vals · CorpFin 65.31%Vals · TaxEval 77.11%Vals · MortgageTax 67.73%Vals · MedQA 92.06%Vals · SWE-bench Verified 77.4% (Mini-SWE-agent)Vals · Vibe Code Bench 55.77% (Claude Code)Vals · Code Migration 39.89%Vals · GPQA Diamond 85.61%Vals · MMLU Pro 87.34%Vals · AIME 92.29%
Enkrypt AI Safety Leaderboard ↗4 measuresread 2026-09-26
Enkrypt · Harmful content risk 0.6%Enkrypt · CBRN risk 15.3%Enkrypt · Bias risk 81.1%Enkrypt · Insecure code risk 32.9%
Vectara · Factual consistency 89.4%

Badge

PublicAI Index badge for Claude Sonnet 4.6[![PublicAI Index](https://publicai.io/model-index/badge?model=claude-sonnet-4-6)](https://publicai.io/model-index/m/claude-sonnet-4-6)