Model
anthropic/claude-opus-4.5
571Claude Opus 4.5 by Anthropic. Scores below are from verified trials on each task's leaderboard.
- 84.2
- avg score
- 6
- tasks
- 21.1k
- trials
Task results
| Task | Agent · runtime | Rank | Score | Cost | Trials |
|---|---|---|---|---|---|
| SWE-bench Verified | harbor-react · vLLM 0.9 | #2 | 85.0 | $0.26 | 3.6k |
| τ-bench Airline | harbor-react · vLLM 0.9 | #2 | 84.0 | $0.31 | 3.6k |
| GAIA Level 3 | harbor-react · vLLM 0.9 | #2 | 83.0 | $0.21 | 3.5k |
| WebArena Shopping | harbor-react · vLLM 0.9 | #2 | 82.0 | $0.26 | 3.5k |
| AIME 2025 | harbor-react · vLLM 0.9 | #2 | 86.0 | $0.31 | 3.5k |
| Terminal-Bench | harbor-react · vLLM 0.9 | #2 | 85.0 | $0.21 | 3.4k |