Claude Sonnet 5vsClaude Sonnet 4.6
Across 5 shared benchmarks, Claude Sonnet 5 leads overall: Claude Sonnet 5 wins 5, Claude Sonnet 4.6 wins 0, with 0 ties and an average score difference of +11.34.
Claude Sonnet 5
Anthropic · 2026-06-30 · Multimodal model
Claude Sonnet 4.6
Anthropic · 2026-02-17 · Chat model
Claude Sonnet 55 wins(100%)(0%)0 winsClaude Sonnet 4.6
Benchmark scores
Grouped by capability, sorted by largest gap within each. 5 shared benchmarks.
Coding and Software Engineer
Claude Sonnet 5 2/2| Benchmark | Claude Sonnet 5 | Claude Sonnet 4.6 | Diff |
|---|---|---|---|
| DeepSWE | 547 / 16Deep Thinking (With Tools) | 3014 / 16Thinking High (With Tools) | +24 |
| SWE-bench Verified | 85.206 / 109极高强度思考（工具） | 79.6017 / 109 | +5.60 |
AI Agent - Information Search
Claude Sonnet 5 1/1| Benchmark | Claude Sonnet 5 | Claude Sonnet 4.6 | Diff |
|---|---|---|---|
| BrowseComp | 84.705 / 49Thinking (With Tools + Internet) | 74.7023 / 49 | +10 |
AI Agent - Tool Usage
Claude Sonnet 5 1/1| Benchmark | Claude Sonnet 5 | Claude Sonnet 4.6 | Diff |
|---|---|---|---|
| OSWorld-Verified | 81.203 / 20极高强度思考（工具） | 72.5012 / 20 | +8.70 |
General Knowledge
Claude Sonnet 5 1/1| Benchmark | Claude Sonnet 5 | Claude Sonnet 4.6 | Diff |
|---|---|---|---|
| HLE | 57.407 / 164极高强度思考（工具） | 4929 / 164 | +8.40 |
Specs
| Field | Claude Sonnet 5 | Claude Sonnet 4.6 |
|---|---|---|
| Publisher | Anthropic | Anthropic |
| Release date | 2026-06-30 | 2026-02-17 |
| Model type | Multimodal model | Chat model |
| Architecture | Dense | Dense |
| Parameters | Not available | Not available |
| Context length | 1M | 1M |
| Max output | 128K | 8K |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | Claude Sonnet 5 | Claude Sonnet 4.6 |
|---|---|---|
| Text input | $2 / 1M tokens | $3 / 1M tokens |
| Text output | $10 / 1M tokens | $15 / 1M tokens |
| Cache read | $0.3 / 1M tokens | $0.3 / 1M tokens |
| Cache write | $3.75 / 1M tokens | $3.75 / 1M tokens |
Summary
- Claude Sonnet 5leads in:Coding and Software Engineer (2/2), AI Agent - Information Search (1/1), AI Agent - Tool Usage (1/1), General Knowledge (1/1)
On average across the 5 shared benchmarks, Claude Sonnet 5 scores 11.34 higher.
Largest single-benchmark gap: DeepSWE — Claude Sonnet 5 54 vs Claude Sonnet 4.6 30 (+24).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.