Editorial analysis: For AI practitioners building agents, lower-cost models that natively handle multi-step tool use shift production tradeoffs toward mid-tier models rather than always using top-tier, expensive models. In a June 30 blog post, Anthropic introduced Claude Sonnet 5, describing it as "the most agentic Sonnet model yet" and an upgrade to the Sonnet line (Anthropic blog). Reporting by TechCrunch and Axios notes that Sonnet 5 approaches the performance of Opus 4.8 while aiming to be cheaper to operate (TechCrunch; Axios). TechCrunch and other outlets report introductory API pricing at $2 per million input tokens and $10 per million output tokens through August 31, moving to $3 per million input tokens thereafter (TechCrunch). The model is available across Anthropic's platform, on AWS Bedrock, and is rolling into GitHub Copilot and other integrations per AWS and GitHub blog posts.
Claude Sonnet 5 – benchmark results