CTBench - Root Cause Identification: leaderboard
CTBench (arXiv 2608.12002v1, Table 5): expert-built telecom network troubleshooting tasks solved by agent harness and model pairs in a sandbox that exposes only permitted diagnostic commands. Intersection over union between predicted and gold root-cause label sets, percent. Higher is better.
Metric: IoU of root-cause label sets (%). Source: arxiv.org. Saturation forecast: Around February 2027. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.5 | 66.83 | #33 |
| 2 | Qwen 3.7 Max | 52.38 | #85 |
| 3 | DeepSeek V4 Pro | 40.08 | #112 |
| 4 | Qwen 3.7 Plus | 38.96 | #98 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=ctbench-root-cause-identification · How It Works · Data refreshed daily, snapshot 2026-11-05.