Claude Sonnet 4.6
Language modelPrevious generationCard summarizedA full upgrade of the model's skills across coding, computer use, long-context reasoning
Active (legacy) · retires not before 17 Feb 2027● Re-checked 12 Oct 2026 · 98 facts · 0 corrections · 3 unconfirmedSource: System Card: Claude Sonnet 4.6 (135 pp)
Use it via API
Where to call it
- Claude API
claude-sonnet-4-6 - Amazon Bedrock
anthropic.claude-sonnet-4-6 - Google Vertex AI
claude-sonnet-4-6 - Microsoft Foundry / Azure
claude-sonnet-4-6
Supports
- Image input
- PDF input
- Structured outputs
- Reasoning control
- Tool calling
- Web search
- Prompt caching
- Batch API
Good for
- Coding and agentic tool use: 79.6% SWE-bench Verified, 59.1% Terminal-Bench 2.0
- Computer use: 72.5% OSWorld-Verified, close to Opus 4.6
- Knowledge work: highest GDPval-AA Elo (1633) in the card's table
Watch out for
Replaced by
Claude Sonnet 5.5$2.00 / $10.00
This model: $3.00 / $15.00 per 1M tokens
Compare side by sideOverview
Sonnet 4.6
A full upgrade of the model's skills across coding, computer use, long-context reasoning
- Cost
- $3.00 in
- Context
- 1M
- Input / output
- $3.00 / $15.00
- Max output
- 128K
- Knows up to
- August 2025
- Released
- 17 Feb 2026
Anthropic's Sonnet-class model released after Claude Opus 4.6. It improves substantially over Claude Sonnet 4.5 across coding, agentic, reasoning and computer-use skills, approaches or matches Opus 4.6 on several evaluations, and is deployed under the ASL-3 Standard. p.3
claude-sonnet-4-6- Input / output per 1M
- $3.00 / $15.00
- Context window
- 1M
- Max output
- 128K
- Knows up to
- August 2025
- Reasoning
- Adaptive (extended deprecated)
- Weights
- Closed
About this card. A model card is the maker's own account of a model: what it's for, how it was trained and tested, and where it falls short. This page condenses Anthropic's System Card: Claude Sonnet 4.6. Every claim carries its source page.