Changes from previous version
Narrows the agentic gap with Opus-class models while staying at Sonnet-tier pricing; strict improvement on BrowseComp and OSWorld-Verified versus Sonnet 4.6, with lower rates of misaligned behavior in safety evaluations.
Release Summary
Anthropic's most agentic Sonnet yet: top-tier coding and tool use at Sonnet pricing with a 1M context window. Performance is close to Claude Opus 4.8 on agentic benchmarks, with a substantial upgrade over Claude Sonnet 4.6 on reasoning, tool use, coding, and knowledge work. Default model on Free and Pro plans and new default in Claude Code for Pro users. Introductory API pricing of $2/M input and $10/M output through August 31, 2026, then standard $3/$15; available as claude-sonnet-5 on the Claude Platform.
Timeline
Claude Sonnet 5 released
Anthropic releases Sonnet 5 across all Claude plans, Claude Code, and the Claude Platform with introductory API pricing and benchmark gains over Sonnet 4.6 approaching Opus 4.8 on agentic tasks.
Major API outage sparks speculation
Claude Code and all API models go down with 500 errors. Twitter speculates imminent Sonnet 5 release due to infrastructure preparation.
Opus 4.5 API errors
Elevated error rates on Opus 4.5 API suggest infrastructure strain, possibly related to new model deployment.
Model ID spotted in Vertex AI
Model identifier 'claude-sonnet-5@20260203' appears in cloud service logs tied to Vertex AI, following Anthropic's naming convention.