Models · The Decoder ·
Anthropic's new Claude Sonnet 5 closes the gap to the pricier Opus model series
Anthropic released Claude Sonnet 5, saying it outperforms Sonnet 4.6 on all benchmarks and slightly exceeds Opus 4.8 on the GDPval-AA v2 knowledge-work test with a score of 1,618. The company also noted that the model scores well below US-government-blocked models on cybersecurity tasks.