Anthropic released Claude Sonnet 5 on June 30 as a more affordable alternative to its larger models. The new version delivers near-Opus performance at lower cost, with introductory pricing at $2 per million input tokens and $10 per million output tokens through August 31.

Performance Meets Affordability

Claude Sonnet 5 closes the gap between mid-tier and high-end models. It performs on par with Claude Opus 4.8 for many tasks, but costs significantly less. The model handles coding, debugging, and multi-step reasoning with confidence previously reserved for larger, pricier systems.

Sonnet 5 supports a 1 million token context window by default, 128k max output tokens, and adaptive thinking—features that let developers handle longer documents and more complex problems without hitting artificial limits.

Who Gets It

Starting immediately, Claude Sonnet 5 is the default model for Free and Pro users on claude.ai. Max, Team, and Enterprise customers also have access. The introductory pricing runs through August 31, 2026. Standard pricing kicks in September 1 at $3 per million input tokens and $15 per million output tokens.

Why It Matters

Sonnet 5 removes friction for developers building agents and automation tools. At the old Sonnet 4.6 price, developers often stepped up to Opus for more demanding tasks. Sonnet 5 splits the difference. It’s agentic enough for complex workflows but affordable enough to run in production at scale.

Anthropic is betting that cheaper, capable models drive higher volume and longer customer relationships.

FYI (keeping you in the loop)

Is Claude Sonnet 5 as smart as Claude Opus 4.8?

Not identically, but very close for most tasks. Sonnet 5 handles code, debugging, and reasoning extremely well. Opus 4.8 still leads on the most complex problems, but the gap is narrow enough that cost becomes the deciding factor for most users.