Claude Sonnet 4.5

Claude Sonnet 4.5 is the best coding model in the world. It's the strongest model for building complex agents. It’s the best model at using computers. And it shows substantial gains in reasoning and math. Highest intelligence across most tasks with exceptional agent and coding capabilities.

Benchmark results

Benchmark Score Tags Source
AIME 2025 87.0% self-reported llm-stats link →
GPQA 83.4% self-reported llm-stats link →
MMMLU 89.1% self-reported llm-stats link →
MMMUval 77.8% self-reported llm-stats link →
OSWorld 61.4% self-reported llm-stats link →
SWE-bench Verified (Agentic Coding) 77.2% self-reported llm-stats link →
TAU-bench Airline 70.0% self-reported llm-stats link →
TAU-bench Retail 86.2% self-reported llm-stats link →
Terminal-Bench 50.0% self-reported llm-stats link →