Claude Opus 5vsOpus 4.7
Across 7 shared benchmarks, Claude Opus 5 leads overall: Claude Opus 5 wins 7, Opus 4.7 wins 0, with 0 ties and an average score difference of +13.37.
Claude Opus 5
Anthropic · 2026-07-24 · Reasoning model
Opus 4.7
Anthropic · 2026-04-16 · Reasoning model
Claude Opus 57 wins(100%)(0%)0 winsOpus 4.7
Benchmark scores
Grouped by capability, sorted by largest gap within each. 7 shared benchmarks.
General Knowledge
Claude Opus 5 4/4| Benchmark | Claude Opus 5 | Opus 4.7 | Diff |
|---|---|---|---|
| ARC-AGI-3 | 30.201 / 9Thinking High (No Tools) | 08 / 9Thinking High (No Tools) | +30.20 |
| ARC-AGI-2 | 90.401 / 62最高（无工具） | 75.8011 / 62最高（无工具） | +14.60 |
| HLE | 64.701 / 172Max (With Tools) | 54.7013 / 172Extended (with tools) | +10 |
| ARC-AGI | 97.501 / 68极高强度思考（无工具） | 93.5011 / 68Thinking High (No Tools) | +4 |
Coding and Software Engineer
Claude Opus 5 2/2| Benchmark | Claude Opus 5 | Opus 4.7 | Diff |
|---|---|---|---|
| SWE-Bench Pro - Public | 79.202 / 54Max (With Tools) | 64.307 / 54Extended (with tools) | +14.90 |
| SWE-bench Verified | 961 / 112Max (With Tools) | 87.606 / 112Extended (with tools) | +8.40 |
AI Agent - Information Search
Claude Opus 5 1/1| Benchmark | Claude Opus 5 | Opus 4.7 | Diff |
|---|---|---|---|
| BrowseComp | 90.802 / 53Max (With Tools + Internet) | 79.3017 / 53Extended (with tools) | +11.50 |
Specs
| Field | Claude Opus 5 | Opus 4.7 |
|---|---|---|
| Publisher | Anthropic | Anthropic |
| Release date | 2026-07-24 | 2026-04-16 |
| Model type | Reasoning model | Reasoning model |
| Architecture | Dense | Dense |
| Parameters | Not available | Not available |
| Context length | 1M | 1000K |
| Max output | 128K | 128K |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | Claude Opus 5 | Opus 4.7 |
|---|---|---|
| Text input | $5 / 1M tokens | $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 | $6.25 / 1M tokens |
Summary
- Claude Opus 5leads in:General Knowledge (4/4), Coding and Software Engineer (2/2), AI Agent - Information Search (1/1)
On average across the 7 shared benchmarks, Claude Opus 5 scores 13.37 higher.
Largest single-benchmark gap: ARC-AGI-3 — Claude Opus 5 30.20 vs Opus 4.7 0 (+30.20).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.