Claude Sonnet 4.6 Thinking
Claude Sonnet 4.6 with extended thinking enabled for tougher coding, planning, and multi‑tool tasks. Ideal for long‑horizon agent workflows and complex problem solving.
- Reasoning
- Vision
- Native PDF input
- Tool Calling
- Structured Output
Added Feb 17, 2026
Pricing
Auto routing · per 1M tokens- Input
- $3.00
- Output
- $15.00
- Cache read
- $0.30
Specifications
- Context window
- 1M
- Max output
- 128K
Benchmarks
Benchmarks
Sourced from Vectara.
Hallucination Rate
10.6%
Factual Consistency
89.4%
Answer Rate
99.9%
Avg Summary Length
Average generated summary length
114.7
Last updated 2026-10-22 · Matched as anthropic/claude-sonnet-4-6
Vectara LeaderboardLeaderboard results
These leaderboards tested Claude Sonnet 4.6, another version of this model.
| Benchmark | Score | Rank on NanoGPT |
|---|---|---|
Epoch AI Benchmarking Hub Epoch AI data, CC BY 4.0 · Updated Nov 12, 2026 | ||
Epoch Capabilities Index | 152.9 | #52 of 115 |
SWE-bench Verified | 75.2% | #12 of 28 |
GPQA Diamond | 87.4% | #59 of 119 |
SimpleQA Verified | 35.5% | #55 of 82 |
OTIS Mock AIME 2024–2025 | 85.8% | #73 of 121 |
Chess Puzzles | 23.0% | #55 of 82 |
Mystery Game Puzzles | 16.0% | #62 of 77 |
EQ-Bench 4 | 1197 | #20 of 31 |
Creative Writing v3 | 1811 | #24 of 96 |
Longform Writing | 79.9 | #16 of 97 |
EQ-Bench 3 | 1710 | #5 of 49 |
UGI Leaderboard UGI Leaderboard data, Apache-2.0 · Updated Oct 16, 2026 | ||
UGI Score | 53.5 | #27 of 149 |
UGI Willingness | 1.8 | #128 of 149 |
UGI Writing | 64.5 | #25 of 149 |
Filtered and ranked to models on NanoGPT.
Providers
Auto routing is available for this model. Explicit provider selection is not available.
| Cache write /M | ||||||
|---|---|---|---|---|---|---|
 Recommended | $3.00 | $15.00 | $0.30 | $3.75 | N/A | N/A |
