ToolMaze (Failure-Aware Prompt) - Perturbed Task Success: leaderboard
Metric: Task success rate (%) averaged over the four tool-failure modes, macro-averaged over the four topology levels; 400 DAG-structured tool-use tasks (four topology levels) over 270 tools, each also run under four injected tool-failure modes (explicit or implicit, transient or permanent); sandboxed, 25 agent steps, temperature 1, a failure-aware prompt that warns of tool failures and outlines recovery strategies (the clean runs keep the standard prompt); higher is better. Source: arxiv.org. Saturation forecast: Around March 2027. 9 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 46.25 |
| 2 | GPT-5.5 | 43.56 |
| 3 | Claude Sonnet 4.6 | 42 |
| 4 | DeepSeek V4 Pro | 40.81 |
| 5 | GLM-5.1 | 33.75 |
| 6 | Qwen 3.6 27B | 23.94 |
| 7 | Qwen 3.5 397B A17B | 22.06 |
| 8 | MiniMax-M2.7 | 21.94 |
| 9 | Qwen 3.5 35B A3B | 19.31 |
Interactive version: theaggregate.ai/benchmark?slug=toolmaze-failure-aware-prompt-perturbed-task-success · How It Works · Data refreshed daily, snapshot 2026-09-29.