Atlas

Models

← All models

Sonnet 4

Anthropic · 2025-05-22

Sonnet 4 is Anthropic's balanced mid-tier Claude 4 model, a hybrid-reasoning model offering standard inference plus optional extended thinking. The successor to Claude 3.7 Sonnet, it is tuned for strong coding, parallel tool use, and general-purpose work.

Benchmark scores

BenchmarkScore
A-Fantasia44.0
A-Fantasia (Valid Attempts with Retries)44.0
A-Fantasia Backwards Spelling32.0
A-Fantasia Chess29.0
A-Fantasia Chess (Valid Attempts with Retries)29.0
A-Fantasia Cube (Valid Attempts with Retries)71.0
A-Fantasia Cube Rotation71.0
A-Fantasia Spelling (Valid Attempts with Retries)32.0
AA-LCR64.7
AA-LCR v1.170.3
AA-Omniscience Index0.2
Active Probing Benchmark (APB)26.4
Active Probing Benchmark (Biology)29.2
Active Probing Benchmark (Chemistry)23.3
Active Probing Benchmark (Computer Science)18.3
Active Probing Benchmark (Economics)42.5
Active Probing Benchmark (Human Disease)35.0
Active Probing Benchmark (Mathematics)17.5
Active Probing Benchmark (Philosophy)21.7
Active Probing Benchmark (Physics)23.3
Agent Red Teaming k=100 (CAIS)86.6
Aider Polyglot61.3
AIME 2024 (avg@8)82.5
AIME 2024-202576.3
AIME 202570.5
AIME 2025 (Artificial Analysis avg@10)74.3
AIME 2025 (avg@8)70.0
ALE-Bench1088
ARC-AGI-1 Semi-Private40.0
ARC-AGI-25.9
Loading Atlas data…