IntelligenceFEB 01
Claude Sonnet 4.5: Benchmarks reveal 24% gain in reasoning capabilities
Anthropic's latest model demonstrates significant improvements in complex problem solving and code generation, challenging current state-of-the-art benchmarks.