Claude Sonnet 5
Anthropic's most capable Sonnet-class model, with adaptive thinking and selectable reasoning effort levels, at introductory pricing below Sonnet 4.6.
Pricing
Input$2 / 1M checked Jul 19, 2026
Cached input$0.20 / 1M
Output$10 / 1M checked Jul 19, 2026
Introductory pricing through Aug 31, 2026; standard $3 in / $15 out from Sep 1, 2026. Batch halves either rate. Uses Anthropic's newer tokenizer, which produces roughly 30% more tokens for the same text than Claude 4.6 and earlier.
Cost bandBalanced
Free tierNo free API tier.
Limits
Context window1M checked Jul 19, 2026
DeploymentClaude Platform, Bedrock, Vertex AI, Foundry
OpenAI-compatible APINo
ModalitiesText, Image input, PDF input
Where it fits
Best forAgents and coding at scale; the default Anthropic pick while introductory pricing lasts.
Business fitUndercuts Sonnet 4.6 ($2/$10 vs $3/$15) through Aug 31, 2026, then matches it.
Capabilities
Adaptive thinking (effort levels)Function callingStructured outputsMCP
Change history
- Jun 30, 2026New modelClaude Sonnet 5 added to the index $2 in / $10 out per 1M introductory through Aug 31, 2026 (then $3/$15), 1M context. Anthropic pricing
Data
Last checkedJul 19, 2026
Cross-checked againstopenrouter.ai/anthropic/claude-sonnet-5
Verify with provider documentation before committing spend. Report an error and it gets fixed with a logged correction.