MODELS
Claude Sonnet 4.6
Anthropic's mid-tier workhorse for coding, agents, and long-context reasoning.
Specs
- Context window
- 200,000
- Max output
- 8,192
- Modalities
- text, image
- Tool use
- ✓
- Vision
- ✓
- Streaming
- ✓
- License
- proprietary
- Released
- 2025-12-01
Pricing
- Input / 1M
- $3.00
- Output / 1M
- $15.00
- Cached input / 1M
- $0.30
Cost estimate
Claude Sonnet 4.6 is Anthropic's mid-tier model, positioned between Haiku and Opus. It handles 200K context, supports tool use and vision, and is tuned for coding, agentic workflows, and structured reasoning. Priced at $3/M input and $15/M output, it targets teams that need Opus-level reliability on most tasks without the Opus bill.
Editor's verdict
Pick Sonnet 4.6 when you're building coding agents or long-document pipelines and want Claude's instruction-following discipline without paying Opus rates. It's the obvious default against GPT-5 and Gemini 2.5 Pro for code-heavy work, though Gemini still wins on raw context length and price. The 8K output cap bites on long refactors — plan to stream and chunk.
Reviews
No reviews yet. Be the first.
Last updated: 2026-04-29