‹ PublicAI Index
The LLM benchmark aggregator.
Claude Sonnet 4
Anthropic
Strongest in Toxicity avoidance (#10 of 268), weakest in Agents (#153 of 298). Above par in 17 of 28 scopes. Among the models it meets almost everywhere, it finishes behind Claude Fable 5 and Claude Fable 5.1 and ahead of Claude 3.5 Sonnet and Grok 3 Mini.
§ 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.
Safety56.7−4.7#28/3293/3
Toxicity avoidance58.7−0.3#10/2681/1
Jailbreak resistance61.9−3.1#23/2681/1
Fairness62.1−7.9#30/2922/2
Safe-prompt compliance55.5−4.9#30/821/1
Factual grounding49.1−21.6#60/1041/1
Secure code59−7.2#71/2701/1
Harm refusal54.3−7.2#82/2922/2
Core abilities56.9−8.4#30/1991/4
General intelligence60−9.4#27/1781/3
Coding52.8−13.4#62/1792/6
Code generation50−15.3#14/201/1
Agentic coding54.1−12.8#47/1731/5
Knowledge51.4−11.1#97/1971/3
Academic knowledge53.3−10.6#67/1231/1
Reasoning48.1−17.3#112/1953/5
Mathematics52.7−11.1#72/1521/2
Science49.2−14.3#79/1221/1
Reasoning44−25#94/1472/3
Professional48.6−13.8#115/1941/2
Legal52.9−13#69/1731/1
Finance48.7−13.7#104/1621/1
Medical45.9−20.6#115/1491/1
Human preference56.9−11.9#130/3541/1
Human preference56.9−11.9#130/3541/1
Agents48.8−17.6#153/2982/6
Computer use49−19.1#36/681/2
Knowledge work47.8−22.9#116/2081/2
§ 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, left for the other.
§ 3 · Sources
Where the numbers come from
12 publications, 30 figures. Every one links to the page it was read from.
LMArena Text 1402
GDPval-AA 698
ARC-AGI-2 5.9%
LiveCodeBench 55.9%
Aider polyglot 61.3%
OSWorld 43.9%
Kagi LLM Benchmark 73%
SimpleBench 45.5%
Vals · LegalBench 82.06%Vals · CorpFin 61.23%Vals · TaxEval 72%Vals · MortgageTax 49.88%Vals · MedQA 92.71%Vals · MedCode 34.96%Vals · MedScribe 69.35%Vals · GPQA Diamond 75%Vals · MMLU Pro 83.86%Vals · AIME 76.25%
HELM Safety · HarmBench 98%HELM Safety · SimpleSafetyTests 100%HELM Safety · Anthropic Red Team 99.2%HELM Safety · BBQ 96.6%HELM Safety · XSTest 96.6%
Enkrypt · Jailbreak risk 2.8%Enkrypt · Harmful content risk 0.6%Enkrypt · CBRN risk 37%Enkrypt · Toxicity risk 0.2%Enkrypt · Bias risk 57.1%Enkrypt · Insecure code risk 14.7%
Vectara · Factual consistency 89.7%
Badge
[](https://publicai.io/model-index/m/claude-sonnet-4)