Anthropic · Claude Sonnet
Claude Sonnet 5
Anthropic’s current Sonnet balance for everyday professional work, coding and agentic tasks, with adaptive thinking by default.
A strong default candidate for professional work, but migration tests must include token usage and corrections, not quality impressions alone.
Use it for
- Professional writing and analysis
- Coding and tool-using workflows
- Long-context work that does not need the top tier
Migration is not a blind model-ID swap if your API depends on manual thinking budgets, sampling controls or tightly sized output limits.
Effort without the jargon
Let the model decide when to reason, then set effort according to consequence rather than prompt length.
Advanced recordCost, limits, privacy and operations
Sonnet is the lower-cost Claude candidate in this set, but long context and large outputs can still create a large bill; pin the model ID and measure total tokens.
- API cost
- Officially disclosedClaude API list price: US$3 per million input tokens and US$15 per million output tokens. Introductory US$2/US$10 pricing is listed through 31 August 2026.
- Context window
- Officially disclosed1,000,000 tokens
- Maximum output
- Officially disclosed128,000 tokens (synchronous Messages API)
- Knowledge cutoff
- Officially disclosedReliable knowledge: January 2026; training data: January 2026
- Recorded inputs
- Text · Images
- Recorded output
- Text
Data handling is a product decision
Depends on surfaceAnthropic says commercial API inputs and outputs are deleted from its backend within 30 days by default, with exceptions for stored services, agreed controls, safety enforcement and law. Consumer Claude and cloud platforms have separate terms.
Operational constraints
- The API and Claude Code default effort to high unless it is set explicitly.
- The dateless API ID is a pinned snapshot, not an evergreen alias.
- A 300,000-token batch output requires a beta header and is not the synchronous default.
Unknown means the cited official record does not disclose a safe value. It is not an estimate. Prices are provider list prices in USD where stated and can change before this record's review date.
Keep the evidence separate
Anthropic presents Sonnet 5 as a capability upgrade over Sonnet 4.6, with its largest gains in coding and agentic tasks.
He surfaced the practical migration details: adaptive thinking, removed sampling controls and a tokenizer that can increase effective cost even when listed per-token pricing is unchanged.
Read the independent source ↗No Human Bit result is claimed; use this as evaluation guidance only.
Not scheduled
No Human Bit result is claimed for this model yet.