Claude Sonnet 4.6vsClaude Sonnet 4
Across 10 shared benchmarks, Claude Sonnet 4.6 leads overall: Claude Sonnet 4.6 wins 9, Claude Sonnet 4 wins 1, with 0 ties and an average score difference of +20.63.
Claude Sonnet 4.6
Anthropic · 2026-02-17 · Chat model
Claude Sonnet 4
Anthropic · 2025-05-23 · Reasoning model
Claude Sonnet 4.69 wins(90%)(10%)1 winClaude Sonnet 4
Benchmark scores
Grouped by capability, sorted by largest gap within each. 10 shared benchmarks.
General Knowledge
Claude Sonnet 4.6 3/3| Benchmark | Claude Sonnet 4.6 | Claude Sonnet 4 | Diff |
|---|---|---|---|
| ARC-AGI-2 | 58.3018 / 59 | 5.9043 / 59 | +52.40 |
| HLE | 4925 / 157 | 9.60134 / 157 | +39.40 |
| GPQA Diamond | 89.9021 / 178 | 83.8057 / 178 | +6.10 |
Agent Level Benchmark
Claude Sonnet 4.6 1/1| Benchmark | Claude Sonnet 4.6 | Claude Sonnet 4 | Diff |
|---|---|---|---|
| τ²-Bench - Telecom | 97.909 / 35 | 6529 / 35 | +32.90 |
AI Agent - Tool Usage
Claude Sonnet 4.6 1/1| Benchmark | Claude Sonnet 4.6 | Claude Sonnet 4 | Diff |
|---|---|---|---|
| OSWorld-Verified | 72.5010 / 18 | 42.2016 / 18 | +30.30 |
Claw-style Agent Evaluation
Claude Sonnet 4.6 1/1| Benchmark | Claude Sonnet 4.6 | Claude Sonnet 4 | Diff |
|---|---|---|---|
| Pinch Bench | 885 / 37Thinking (With Tools) | 80.5022 / 37Thinking (With Tools) | +7.50 |
Coding and Software Engineer
Claude Sonnet 4 1/1| Benchmark | Claude Sonnet 4.6 | Claude Sonnet 4 | Diff |
|---|---|---|---|
| SWE-bench Verified | 79.6017 / 108 | 80.2013 / 108 | -0.60 |
Long Context
Claude Sonnet 4.6 1/1| Benchmark | Claude Sonnet 4.6 | Claude Sonnet 4 | Diff |
|---|---|---|---|
| AA-LCR | 711 / 13 | 6510 / 13 | +6 |
Math and Reasoning
Claude Sonnet 4.6 1/1| Benchmark | Claude Sonnet 4.6 | Claude Sonnet 4 | Diff |
|---|---|---|---|
| FrontierMath - Tier 4 | 8.3034 / 80Thinking (No Tools, 16K Budget) | 072 / 80Normal (No Tools) | +8.30 |
Productivity Knowledge
Claude Sonnet 4.6 1/1| Benchmark | Claude Sonnet 4.6 | Claude Sonnet 4 | Diff |
|---|---|---|---|
| GDPval-AA | 5711 / 21 | 3319 / 21 | +24 |
Specs
| Field | Claude Sonnet 4.6 | Claude Sonnet 4 |
|---|---|---|
| Publisher | Anthropic | Anthropic |
| Release date | 2026-02-17 | 2025-05-23 |
| Model type | Chat model | Reasoning model |
| Architecture | Dense | Dense |
| Parameters | Not available | Not available |
| Context length | 1M | 200K |
| Max output | 8K | 64K |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | Claude Sonnet 4.6 | Claude Sonnet 4 |
|---|---|---|
| Text input | $3 / 1M tokens | Not public |
| Text output | $15 / 1M tokens | Not public |
| Cache read | $0.3 / 1M tokens | Not public |
| Cache write | $3.75 / 1M tokens | Not public |
One or both models have incomplete public pricing.
Summary
- Claude Sonnet 4.6leads in:General Knowledge (3/3), Agent Level Benchmark (1/1), AI Agent - Tool Usage (1/1), Claw-style Agent Evaluation (1/1), Long Context (1/1), Math and Reasoning (1/1), Productivity Knowledge (1/1)
- Claude Sonnet 4leads in:Coding and Software Engineer (1/1)
On average across the 10 shared benchmarks, Claude Sonnet 4.6 scores 20.63 higher.
Largest single-benchmark gap: ARC-AGI-2 — Claude Sonnet 4.6 58.30 vs Claude Sonnet 4 5.90 (+52.40).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.