Claude Opus 4.7: benchmark results
Anthropic's flagship Opus model for complex reasoning, coding, and writing tasks. Provider: Anthropic. Released 2026-04-16. Access: API.
Aggregate Rating 1824 ± 3, rank #49 of 1656 rated models, from 2129 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| 1GC-7RC: One Graphic Card -- Seven Research Challenges! How Good Are AI Agents at Doing Your Job? | 0.3 | 1GC-7RC ($\uparrow$ Agg) | 100 |
| 3DCodeBench - Executability | 91 | Executability (%; Blender 5.0 scripts that run and yield a m | 100 |
| AARRI-Bench (Claude Code) | 62.2 | Task success rate (%) under the 0/1 reward (a task counts on | 100 |
| AARRI-Bench (Claude Code) - Context | 55.9 | Task success rate (%) under the 0/1 reward on the context ta | 100 |
| AARRI-Bench (Claude Code) - Hands-on | 57.1 | Task success rate (%) under the 0/1 reward on the hands-on t | 100 |
| AARRI-Bench (Claude Code) - Mindset | 76.9 | Task success rate (%) under the 0/1 reward on the mindset ta | 100 |
| AARRI-Bench (Hermes Agent) - Mindset | 76.9 | Task success rate (%) under the 0/1 reward on the mindset ta | 100 |
| AARRI-Bench (Mini-SWE-Agent) | 68.3 | Task success rate (%) under the 0/1 reward (a task counts on | 100 |
| AARRI-Bench (Mini-SWE-Agent) - Context | 64.7 | Task success rate (%) under the 0/1 reward on the context ta | 100 |
| AARRI-Bench (Mini-SWE-Agent) - Interaction | 76.2 | Task success rate (%) under the 0/1 reward on the interactio | 100 |
| ADeptS-Bench - Attack Success (Desktop) | 27.2 | Attack success rate (%, lower is better; share of the 501 ma | 100 |
| ADeptS-Bench - Attack Success (Mobile) | 27.7 | Attack success rate (%, lower is better; share of the 358 ma | 100 |
Interactive version: theaggregate.ai/model?slug=claude-opus-4-7 · How It Works · Data refreshed daily, snapshot 2026-10-20.