- 1M token context window
- SWE-Bench Verified 79.6% (vs Opus 4.6 80.8%)
- Cheaper than Opus
- much closer in capability
- matches Opus 4.6 on safety
- SOTA on financial analysis and office tasks.
This is opus level intelligence at sonnet price.
Claude@claudeai
This is Claude Sonnet 4.6: our most capable Sonnet model yet. It’s a full upgrade across coding, computer use, long-context reasoning, agent planning, knowledge work, and design. It also features a 1M token context window in beta.
English
