Atlas

Benchmarks

← All benchmarks

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)

ModelScore
Claude Fable 551.1
Gemini 3.5 Flash50.1
Claude Opus 547.5
Muse Spark 1.147.2
GPT-5.546.6
Kimi K345.3
Qwen3.7-Max45.0
Sonnet 544.7
Opus 4.844.0
GPT-5.6 Sol42.9
GLM-5.242.7
Gemini 3.1 Pro Preview42.2
Grok 4.540.8
Qwen3.7-Plus40.6
DeepSeek-V4-Pro40.4
Kimi K2.7 Code40.2
DeepSeek-V4-Flash39.6
Kimi K2.638.5
Inkling38.1
GPT-5.4 Mini34.6
Mistral Medium 3.533.7
MiniMax M332.1
GPT-5.4 Nano31.5
Qwen3.5 397B A17B30.9
Haiku 4.529.4
Step 3.7 Flash29.1
Nemotron 3 Ultra 550B A55B28.9
Gemma 4 31B IT28.3
Gemini 3.1 Flash-Lite Preview28.0
Grok 4.326.3
gpt-oss-120b25.5
Loading Atlas data…