claude-opus-4-7
1,298±7
1,292CI1,305
aiqbench · LLM benchmarks
Benchmarks
- Sources
- 8
- Entries tracked
- 22,308
- Last sync
- 24 min24 minutes ago
All sources syncing normally
- Arena · Text· 12 hours ago
- Arena · WebDev· 12 hours ago
- Arena · Vision· 12 hours ago
- Arena · Agent· 12 hours ago
- GPQA Diamond· 24 minutes ago
- SWE-bench· 24 minutes ago
- Terminal-Bench· 24 minutes ago
- ARC-AGI-2· 24 minutes ago
Filter by org:
Released:
One grid, every column. Pick axes in the Axes panel — one axis ranks like a single board, several blend into a weighted composite, all of them read as the index. Rank is assigned before any filter and never renumbers.
| Model | Org | ± CI | Swing | Add to compare | |||||
|---|---|---|---|---|---|---|---|---|---|
| 4 | claude-opus-4-7 | anthropic | Sep 2026 | $5 / $25 | 1,298 | 1,292–1,305 | #1–18 |
Showing Vision Arena exactly as the source published it — its own score and confidence interval, no blending. Select a second axis to build a weighted composite.