Agentic Tool-use on τ2-Bench (Retail and Telecom)
85.79Overall Success Rate
Claude-Opus-4.5
Updated 1mo ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2026.01 | 85.79 | 80.88 | 90.7 | |
2026.01 | 83.69 | 77.72 | 89.65 | |
2026.01 | 76.58 | 77.19 | 75.96 | |
2026.01 | 69.63 | 78.95 | 60.31 | |
2026.01 | 69.52 | 68.64 | 70.39 | |
2026.01 | 63.7 | 75.2 | 52.19 | |
2026.01 | 60.31 | 72.15 | 48.46 | |
2026.01 | 58.3 | - | - | |
2026.01 | 57.69 | 68 | 47.37 | |
2026.01 | 54.58 | 71.26 | 37.89 | |
2026.01 | 54 | 74 | 34 | |
2026.01 | 53.16 | 69.12 | 37.19 | |
2026.01 | 49.7 | 64.7 | 34.7 | |
2026.01 | 46.05 | 55 | 37.1 | |
2026.01 | 44 | - | - | |
| 32.5 | - | - | ||
2026.01 | 16.7 | - | - |