TeleCom-Bench - Root Cause Diagnosis: leaderboard
Knowledge-application task of TeleCom-Bench (Xiao et al., ZTE Corporation, KDD 2026) mined from live network-operation agent trajectories: the model deduces the root cause of a network fault from alarms and performance data (the mean of a fault-agent set and a single-domain alarm-analysis set). Exact match in percent, with reasoning enabled, temperature 0.7 and a majority vote over three samples; higher is better.
Metric: Exact match (%; mean of a fault-agent set and a single-domain alarm-analysis set). Source: arxiv.org. Saturation forecast: Around February 2028. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 235B A22B (Thinking) | 71.49 | #622 (Qwen 3 235B A22B) |
| 2 | DeepSeek V3.2 (Thinking) | 63 | #403 (DeepSeek V3.2) |
| 3 | Qwen 3 32B (Thinking) | 60.92 | #960 (Qwen 3 32B) |
| 4 | GLM-4.7 (Thinking) | 26.63 | #365 (GLM-4.7) |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=telecom-bench-root-cause-diagnosis · How It Works · Data refreshed daily, snapshot 2026-11-12.