Claude Opus 4.7 benchmark ranking on every AI leaderboard
closed weights · #8 overall, podium score 20.7 · top 10 on 5 of 10 · released Apr 16, 2026
10 of 10 boards
| Rank | Board | Score | |||
|---|---|---|---|---|---|
| 3 |  Text ArenaChat · as “claude-opus-4-7-high” | Chat | 1,502 | 100 | |
| 3 | Vision ArenaVision · as “claude-opus-4-7-high” | Vision | 1,301 | 100 | |
| 4 | Search ArenaAgents and tools · as “claude-opus-4-7” | Agents and tools | 1,233 | 33 | |
| 5 | Document ArenaVision · as “claude-opus-4-7” | Vision | 1,495 | 38 | |
| 9 |  HLEReasoning · as “claude-opus-4-7” | Reasoning | 36.2% | 42 | |
| 13 | MCP AtlasAgents and tools · as “claude-opus-4-7 (max)” | Agents and tools | 79.1% | 34 | |
| 19 |  Epoch ECIReasoning · as “Claude Opus 4.7” | Reasoning | 156.4 | 100 | |
| 24 | WebDev ArenaCoding · as “claude-opus-4-7” | Coding | 1,557 | 100 | |
| 25 |  LiveBenchReasoning · as “claude-opus-4-7-xhigh-effort” | Reasoning | 76.5% | 59 | |
| 32 | GPQA DiamondReasoning · as “claude-opus-4-7_xhigh” | Reasoning | 90.2% | 100 |
Category ranks and more models
Scores as published by each board on the capture date. Model names and logos belong to their owners; logos via logo.dev.