Agentic benchmark
AA ITBench leaderboard
Artificial Analysis ITBench-AA. Every model the catalog carries a published AA ITBench value for, ranked by that value.
CategoryAgentic
MeasureTask success rate
TasksIT incident-response tasks
DifficultyEnterprise IT operations
An independently evaluated IT-operations benchmark from Artificial Analysis.
AA ITBench ranking
10 models with a published AA ITBench value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | Task success rate |
|---|---|---|---|
| 1 |  GPT-5.6 Sol | OpenAI | 56.2 |
| 2 | GPT-5.6 Terra | OpenAI | 51.0 |
| 3 |  Kimi K3 | Moonshot AI | 47.7 |
| 4 |  Claude Opus 4.7 (Adaptive) | Anthropic | 46.7 |
| 5 | GPT-5.5 | OpenAI | 45.8 |
| 6 |  GLM-5.2 | Z.AI | 42.7 |
| 7 |  Qwen3.7 Max | Alibaba | 42.5 |
| 8 |  Gemini 3.5 Flash | 40.3 | |
| 8 | GPT-5.6 Luna | OpenAI | 40.3 |
| 10 | GPT-OSS 120B | OpenAI | 5.6 |
Evidence key: Observed
Rows are ordered by the value Artificial Analysis ITBench-AA Benchmark Leaderboard published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.