Robustness on τ-bench (airline, clean)
How well does the agent maintain accuracy when inputs are perturbed (faults, structural changes, prompt rewording)? See full definition →
Sub-metric Comparison
Baseline vs Perturbed Accuracy
Agent Leaderboard — Robustness
| # | Agent | Acc | Reliability | Robustness Agg | Fault | Struct | Prompt |
|---|---|---|---|---|---|---|---|
| 1 |  Claude Sonnet 4.5 |
78.5% | 0.85 | 0.99 | 1.00 | 1.00 | 0.98 |
| 2 |  Gemini 2.0 Flash |
44.6% | 0.70 | 0.98 | 0.98 | 1.00 | 0.98 |
| 3 | Gemini 3.0 Pro |
80.8% | 0.85 | 0.98 | 1.00 | 1.00 | 0.95 |
| 4 | Claude 3.7 Sonnet |
56.2% | 0.75 | 0.97 | 1.00 | 1.00 | 0.91 |
| 5 |  GPT-5.2 (xhigh) |
67.7% | 0.81 | 0.96 | 1.00 | 1.00 | 0.89 |
| 6 | Gemini 2.5 Flash |
65.4% | 0.73 | 0.95 | 1.00 | 1.00 | 0.86 |
| 7 | GPT-5.2 |
59.2% | 0.80 | 0.95 | 1.00 | 1.00 | 0.84 |
| 8 | Claude Opus 4.5 |
83.1% | 0.88 | 0.94 | 0.97 | 0.93 | 0.93 |
| 9 | O1 |
72.3% | 0.80 | 0.93 | 0.95 | 1.00 | 0.85 |
| 10 | Gemini 2.5 Pro |
73.8% | 0.79 | 0.93 | 0.98 | 0.83 | 0.97 |
| 11 | GPT-4 Turbo |
50.0% | 0.71 | 0.92 | 0.95 | 0.85 | 0.95 |
| 12 | GPT-4o Mini |
32.1% | 0.69 | 0.91 | 1.00 | 0.72 | 1.00 |
| 13 | Claude 3.5 Haiku |
42.3% | 0.68 | 0.85 | 0.81 | 1.00 | 0.74 |