ITBench-AA — leaderboard

Artificial Analysis implementation of IBM's ITBench SRE benchmark, testing AI agents on Kubernetes incident root-cause analysis from offline incident snapshots.

Metric: Average Precision at Full Recall (self-reported). Source: benchmarklist.com. Status: saturation imminent. 23 models tracked.

Top models

#ModelScore
1Claude Opus 4.7 (Max)46.7
2GPT-5.545.8
3Qwen 3.7 Max (Max)42.5
4GLM-5.140.3
5Gemini 3.5 Flash40.3
6Claude Sonnet 4.6 (Max)39.8
7DeepSeek V4 Pro (Max)38.3
8MiMo-V2.5-Pro38.2
9Gemma 4 31B37.3
10Qwen 3.5 27B35.5
11GPT-5.4 Mini35.2
12GPT-5.434.5
13Qwen 3.5 397B A17B34.1
14Grok 4.332.7
15DeepSeek V4 Flash (Max)31.5

Interactive version: theaggregate.ai/benchmark?slug=itbench-aa · How the rankings work · Data refreshed daily, snapshot 2026-07-22.