Anthropic Releases Claude Sonnet 5 with Expanded Agentic Capabilities

Anthropic announced in a press release the launch of Claude Sonnet 5, describing it as its most agentic Sonnet model so far. The new version can plan, use tools such as browsers and terminals, and operate autonomously at levels that previously required larger models. It is now the default model for Free and Pro users and is also available to Max, Team, and Enterprise customers.

Claude Sonnet 5 narrows the performance gap with the higher tier Opus 4.8 model while maintaining lower costs. It shows substantial improvements over Sonnet 4.6 in reasoning, coding, and knowledge tasks. Introductory pricing is set at 2 dollars per million input tokens and 10 dollars per million output tokens until August 31, 2026, after which pricing will move to 3 dollars and 15 dollars respectively.

Safety evaluations indicate that Sonnet 5 has lower rates of undesirable or misaligned behavior than its predecessor and includes cyber safeguards enabled by default. The model is part of Anthropic’s Cyber Verification Program and is available through the Claude API, Claude Code, and the Claude Platform. According to testing data, Sonnet 5 performs better than Sonnet 4.6 on complex agentic tasks while retaining a lower cybersecurity risk profile than the Opus and Mythos series.