What Claude Sonnet 4.6 can do and the modalities it supports.
Performance scores for Claude Sonnet 4.6 across standard benchmarks.
| Benchmark | Score | Date |
|---|---|---|
| SWE-bench Verified | 79.6 | February 17, 2026 |
| OSWorld-Verified | 72.5 | February 17, 2026 |
| Terminal-Bench 2.0 | 59.1 | February 17, 2026 |
| ARC-AGI-2 | 58.3 | February 17, 2026 |
| GDPval-AA Elo | 1633 | February 17, 2026 |
| Finance Agent | 63.3 | February 17, 2026 |
| Humanity's Last Exam | 19.1 | February 17, 2026 |
Token pricing for Claude Sonnet 4.6 API usage.
$3.00/M
per million tokens
$15.00/M
per million tokens
Batch API: 50% discount. Prompt caching: up to 90% savings. 1M context window available in beta with premium long-context rates for requests exceeding 200K tokens.
Alternative models that compete with Claude Sonnet 4.6.
| Model | Family | Context Window | Price (In / Out per M) | Status |
|---|---|---|---|---|
| Gemini 2.5 Pro | Gemini | 1.0M | $1.25/M / $10.00/M | Current |
| GPT-5.4 | GPT | 1.1M | $2.50/M / $15.00/M | Current |