Claude Opus 4.7: benchmarks and ranking
Claude Opus 4.7 (Anthropic) — overall rank: #42 / 153, overall score: 72.5 (90% interval 66.8–77.5), leaderboards: 6. Released: 2026-04-16 · Price (in/out): $5 / $25
Rank on each leaderboard
| Leaderboard | Rank | Published value |
|---|---|---|
| Artificial Analysis | Not rated | — |
| LiveBench | #28 / 64 | 76.5 |
| Epoch AI | #32 / 230 | 157.49 |
| ARC Prize (ARC-AGI) | #26 / 95 | 75.8 |
| Scale AI SEAL | #10 / 44 | 36.2 |
| Vals AI | Not rated | — |
| Terminal-Bench | Not rated | — |
| EQ-Bench | #7 / 32 | 1293.2 |
| SimpleBench | #38 / 112 | 62.9 |
| BenchLM.ai | Not rated | — |
Domain scores
- Coding: 71.2
- Agents: 60.8
- Reasoning: 75.6
- Math: 74.5
- Knowledge: 83.3
- Writing: 69.2
- Instructions: 58.6
- Chat & EQ: 84.0
- Business: 86.8
- Japanese: 83.2
- Vision: 74.3
- Long context: 86.9