EnterpriseOps-Gym-AA
Agents · 2026-07-08
Artificial Analysis' independent implementation of ServiceNow's EnterpriseOps-Gym evaluates LLM agents on 1,117 distinct oracle-mode tasks across eight business domains, with three runs per task (3,351 trajectories). The headline score averages final-state task success across those repeated runs.
Top models (higher is better)
| Model | Score |
|---|---|
| Gemini 3.7 Flash | 50.4 |
| Gemini 3.5 Flash | 50.1 |
| DeepSeek V4 Pro 0813 | 49.6 |
| Grok 4.6 | 48.3 |
| Qwen3.8 Max (0902) | 47.6 |
| Claude Opus 5 | 47.5 |
| Qwen3.8 2.4T A95B | 47.4 |
| Muse Spark 1.2 | 47.3 |
| Muse Spark 1.1 | 47.2 |
| Step 5 Preview | 47.2 |
| DeepSeek V4 Flash Vision Exp | 47.1 |
| GPT-5.5 | 46.6 |
| Kimi K3 | 45.3 |
| Qwen3.7-Max (source unspecified) | 45.0 |
| Sonnet 5 | 44.7 |
| Qwen3.8 27B | 44.2 |
| Opus 4.8 | 44.0 |
| GPT-5.6 Sol | 42.9 |
| GLM-5.2 | 42.7 |
| Inkling-Small | 42.7 |
| Gemini 3.5 Flash-Lite | 42.3 |
| Gemini 3.1 Pro Preview | 42.2 |
| GPT-5.6 Luna | 40.8 |
| Grok 4.5 | 40.8 |
| Qwen3.7-Plus (source unspecified) | 40.6 |
| DeepSeek-V4-Pro | 40.4 |
| Kimi K2.7 Code | 40.2 |
| DeepSeek-V4-Flash | 39.6 |
| Kimi K2.6 | 38.5 |
| GPT-5.6 Terra | 38.5 |
| Inkling | 38.0 |
| GLM-5.3 | 36.4 |
| Muse Glimmer 30B | 34.7 |
| GPT-5.4 Mini | 34.6 |
| Mistral Medium 3.5 | 33.7 |
| GLM-5.3 Flash | 33.2 |
| MiniMax M3 | 32.1 |
| GPT-5.4 Nano | 31.5 |
| Qwen3.5 397B A17B | 30.9 |
| Haiku 4.5 | 29.4 |