Automated page. Written and translated automatically by AI and checked against the sources by software; no human editor. SI-IQ is an index built from public benchmark results, not an IQ test. How it works

How smart is Claude Sonnet 4?

97 SI-IQ · 90% interval 93–101
Rank 70 of 80 ranked models · Plausible rank range 66–73

Claude Sonnet 4 is a model from Anthropic, released on May 22, 2025. Its SI-IQ is 97 (90% interval 93–101), estimated from 5 benchmark results by 3 independent evaluators.

Facts

Developer: Anthropic
Released: May 22, 2025
Access: API access
Benchmarks used: 5 results from 3 evaluators

Ability by area (SI-IQ, 90% interval)

Every benchmark result we used

Each score is one value exactly as the source published it: for Epoch AI the best of its run settings, for OpenRouter the median hosting provider (the developer's own hosting when it ran the test). “Implied SI-IQ” is the ability that this single result points to; the overall SI-IQ is their precision-weighted average.

Benchmark	Area	Score	Implied SI-IQ	Evaluated by	As listed by the source	Retrieved	Source
MATH Level 5	Math	84.4%	89	Epoch AI	`claude-sonnet-4-20250514`	Oct 18, 2026	Epoch AI Benchmarking Hub · data file
OTIS Mock AIME 2024–2025	Math	71.1%	103	Epoch AI	`claude-sonnet-4-20250514_32K`	Oct 18, 2026	Epoch AI Benchmarking Hub · data file
Aider Polyglot	Coding	61.3%	103	Aider	`claude-sonnet-4-20250514 (32k thinking)`	Oct 18, 2026	Aider LLM Leaderboards · data file
GPQA Diamond	Knowledge	79.2%	106	Epoch AI	`claude-sonnet-4-20250514_59K`	Oct 18, 2026	Epoch AI Benchmarking Hub · data file
GPQA Diamond (OpenRouter run)	Knowledge	76.7%	106	OpenRouter	`anthropic/claude-4-sonnet-20250522` Google · median over providers (2 of 2 rows), lower median	Oct 18, 2026	OpenRouter benchmark datasets · data file
τ²-bench Airline (OpenRouter run)	Agents	64.0%	110	OpenRouter	`anthropic/claude-4-sonnet-20250522` Amazon Bedrock · median over providers (2 of 2 rows), lower median	Oct 18, 2026	OpenRouter benchmark datasets · data file

Human preference (Arena)

On the Arena leaderboard, where people vote between two anonymous answers, Claude Sonnet 4 has a rating of 1,402 (1,397–1,406) from 34,132 votes (published Oct 15, 2026). Human preference is a different measure from benchmark ability and is not part of SI-IQ. Arena (Hugging Face)

Models with similar SI-IQ

Model	SI-IQ
Claude Opus 4Anthropic	98	Claude Sonnet 4 vs Claude Opus 4
Claude 3.7 SonnetAnthropic	99	Claude Sonnet 4 vs Claude 3.7 Sonnet
Qwen3-32BAlibaba	95	Claude Sonnet 4 vs Qwen3-32B
GPT-4.1 miniOpenAI	91	Claude Sonnet 4 vs GPT-4.1 mini
GPT-4.1OpenAI	91	Claude Sonnet 4 vs GPT-4.1

Computed with SI-IQ method version 1.0.0. Read the full methodology

Data sources: Aider polyglot leaderboard data (Apache-2.0); Arena leaderboard dataset on Hugging Face (CC BY 4.0); Epoch AI, ‘Capabilities & benchmarking’ (CC BY 4.0); Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter..