Opus 4.7vsClaude Opus 4.6
Across 15 shared benchmarks, Opus 4.7 leads overall: Opus 4.7 wins 11, Claude Opus 4.6 wins 2, with 2 ties and an average score difference of +1.83.
Opus 4.7
Anthropic · 2026-04-16 · Reasoning model
Claude Opus 4.6
Anthropic · 2026-02-05 · Reasoning model
Opus 4.711 wins(73%)Ties2(13%)2 winsClaude Opus 4.6
Benchmark scores
Grouped by capability, sorted by largest gap within each. 15 shared benchmarks.
General Knowledge
Opus 4.7 6/7| Benchmark | Opus 4.7 | Claude Opus 4.6 | Diff |
|---|---|---|---|
| ARC-AGI-2 | 75.8011 / 62最高（无工具） | 66.3017 / 62Extended (no tools) | +9.50 |
| GPQA Diamond | 94.204 / 187Extended (no tools) | 91.3115 / 187Extended (no tools) | +2.89 |
| HLE | 54.7013 / 172Extended (with tools) | 5318 / 172Extended (with tools, internet) | +1.70 |
| ARC-AGI | 93.5011 / 68Thinking High (No Tools) | 9213 / 68Extended (no tools) | +1.50 |
| LiveBench | 76.917 / 115Deep Thinking (No Tools) | 76.338 / 115Thinking High (No Tools) | +0.58 |
| MMLU | 91.506 / 66Normal (No Tools) | 91.057 / 66Extended (no tools) | +0.45 |
| ARC-AGI-3 | 08 / 9Thinking High (No Tools) | 04 / 9最高（无工具） | — |
AI Agent - Tool Usage
Opus 4.7 3/3| Benchmark | Opus 4.7 | Claude Opus 4.6 | Diff |
|---|---|---|---|
| OSWorld-Verified | 7810 / 24Extended (with tools) | 72.7015 / 24Extended (with tools) | +5.30 |
| Terminal Bench 2.0 | 69.406 / 47Extended (with tools) | 65.4011 / 47Extended (with tools) | +4 |
| MCP-Atlas | 79.107 / 27Deep Thinking (With Tools) | 76.8010 / 27Deep Thinking (With Tools) | +2.30 |
Math and Reasoning
Opus 4.7 1/2| Benchmark | Opus 4.7 | Claude Opus 4.6 | Diff |
|---|---|---|---|
| FrontierMath | 43.806 / 60极高强度思考（无工具） | 40.707 / 60最高（无工具） | +3.10 |
| FrontierMath - Tier 4 | 22.9012 / 80极高强度思考（无工具） | 22.9012 / 80最高（无工具） | — |
AI Agent - Information Search
Claude Opus 4.6 1/1| Benchmark | Opus 4.7 | Claude Opus 4.6 | Diff |
|---|---|---|---|
| BrowseComp | 79.3017 / 53Extended (with tools) | 8411 / 53Thinking (With Tools + Internet) | -4.70 |
Coding and Software Engineer
Opus 4.7 1/1| Benchmark | Opus 4.7 | Claude Opus 4.6 | Diff |
|---|---|---|---|
| SWE-bench Verified | 87.606 / 112Extended (with tools) | 80.8410 / 112Extended (with tools) | +6.76 |
Commonsense Reasoning
Claude Opus 4.6 1/1| Benchmark | Opus 4.7 | Claude Opus 4.6 | Diff |
|---|---|---|---|
| Simple Bench | 61.7013 / 63Normal (No Tools) | 67.608 / 63Normal (No Tools) | -5.90 |
Specs
| Field | Opus 4.7 | Claude Opus 4.6 |
|---|---|---|
| Publisher | Anthropic | Anthropic |
| Release date | 2026-04-16 | 2026-02-05 |
| Model type | Reasoning model | Reasoning model |
| Architecture | Dense | Dense |
| Parameters | Not available | Not available |
| Context length | 1000K | 1000K |
| Max output | 128K | 64K |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | Opus 4.7 | Claude Opus 4.6 |
|---|---|---|
| Text input | $5 / 1M tokens | $0.5 / 1M tokens |
| Text output | $25 / 1M tokens | $25 / 1M tokens |
| Cache read | $0.5 / 1M tokens | $0.5 / 1M tokens |
| Cache write | $6.25 / 1M tokens | $10 / 1M tokens |
Summary
- Opus 4.7leads in:General Knowledge (6/7), AI Agent - Tool Usage (3/3), Math and Reasoning (1/2), Coding and Software Engineer (1/1)
- Claude Opus 4.6leads in:AI Agent - Information Search (1/1), Commonsense Reasoning (1/1)
On average across the 15 shared benchmarks, Opus 4.7 scores 1.83 higher.
Largest single-benchmark gap: ARC-AGI-2 — Opus 4.7 75.80 vs Claude Opus 4.6 66.30 (+9.50).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.