Agentic · tools
EnterpriseOps-Gym leaderboard
As of 2026-10-07, the highest measured score on EnterpriseOps-Gym is 51.1% by Claude Fable 5 (with fallback). 29 more models have estimated scores, calibrated from the benchmarks they were measured on.
Measured scores: artificial-analysis.
| # | Model | Score | Source |
|---|---|---|---|
| 1 | Gemini 3.7 Flash (high) | 56.7% | estimated ± 1.8 pp, low confidence |
| 2 | Claude Fable 5.1 (max with fallback) | 55.4% | estimated ± 1.8 pp, low confidence |
| 3 | Claude Sonnet 5.5 (max with fallback) | 55.4% | estimated ± 1.8 pp, low confidence |
| 4 | Claude Opus 5.5 (max with fallback) | 54.8% | estimated ± 1.8 pp, low confidence |
| 5 | Claude Opus 5 (max) | 53.6% | estimated ± 1.8 pp, low confidence |
| 6 | GPT-6 Astra (max) | 52.2% | estimated ± 1.8 pp, low confidence |
| 7 | GPT-6.1 Sol (max) | 51.6% | estimated ± 1.8 pp, low confidence |
| 8 | GPT-5.5 (xhigh) | 51.6% | estimated ± 1.8 pp, low confidence |
| 9 | Claude Fable 5 (with fallback) | 51.1% | measured |
| 10 | Gemini 3.7 Flash (medium) | 50.4% | measured |
| 11 | GPT-5.6 Sol (max) | 50.4% | estimated ± 1.8 pp, medium confidence |
| 12 | Gemini 3.5 Flash (high) | 50.1% | measured |
| 13 | Muse Spark 1.3 (xhigh) | 50.0% | estimated ± 4.4 pp, low confidence |
| 14 | DeepSeek V4 Pro 0813 (max) | 49.6% | measured |
| 15 | Grok 4.6 (high) | 48.3% | measured |
| 16 | Qwen3.8 Max (0902) | 47.6% | measured |
| 17 | Step 5 Preview | 47.2% | measured |
| 18 | Claude Fable 5.1 (xhigh with fallback) | 46.0% | estimated ± 4.4 pp, medium confidence |
| 19 | Claude Fable 5.1 (high with fallback) | 45.5% | estimated ± 4.4 pp, medium confidence |
| 20 | Kimi K3 (max) | 45.3% | measured |
| 21 | Qwen3.8 27B (xhigh) | 44.2% | measured |
| 22 | Qwen3.8 Max | 42.9% | estimated ± 5.3 pp, low confidence |
| 23 | Muse Spark 1.3 (max) | 42.8% | estimated ± 5.3 pp, medium confidence |
| 24 | Qwen3.8 2.4T A95B | 42.6% | estimated ± 5.3 pp, medium confidence |
| 25 | Gemini 4 Argon (high) | 42.5% | estimated ± 6.2 pp, low confidence |
| 26 | DeepSeek V4.1 Flash (max) | 42.4% | estimated ± 6.2 pp, low confidence |
| 27 | GPT-6 Astra (xhigh) | 42.4% | estimated ± 6.2 pp, low confidence |
| 28 | Grok 4.6 (xhigh) | 42.4% | estimated ± 6.2 pp, low confidence |
| 29 | GPT-6 Astra (high) | 42.3% | estimated ± 6.2 pp, low confidence |
| 30 | GPT-6.1 Sol (xhigh) | 42.3% | estimated ± 6.2 pp, low confidence |
| 31 | Grok 4.7 (xhigh) | 42.3% | estimated ± 6.2 pp, low confidence |
| 32 | Qwen3.8 27B (medium) | 42.3% | estimated ± 5.3 pp, medium confidence |
| 33 | GPT-6 Sol (max) | 42.3% | estimated ± 6.2 pp, low confidence |
| 34 | Mistral Large 4 Preview | 42.3% | estimated ± 6.2 pp, low confidence |
| 35 | MiMo-V2.6-Pro | 42.2% | estimated ± 6.2 pp, low confidence |
| 36 | GPT-6 Luna (max) | 42.1% | estimated ± 6.2 pp, low confidence |
| 37 | Gemini 3.8 Flash (high) | 41.9% | estimated ± 5.3 pp, medium confidence |
| 38 | K2 Horizon 375B A23B | 39.5% | estimated ± 5.3 pp, medium confidence |
| 39 | Inkling (xhigh) | 38.0% | measured |
| 40 | GLM-5.3 (max) | 36.4% | measured |
| 41 | Muse Glimmer (high) | 34.7% | measured |
| 42 | GLM-5.3-Flash | 33.2% | measured |
| 43 | MiniMax-M3 | 32.1% | measured |
| 44 | Nemotron 3 Ultra | 28.9% | measured |
All leaderboards
Agentic · terminal
Agentic · tools
- AA-AnalystAgent
- APEX-Agents
- AutomationBench
- EnterpriseOps-Gym
- Harvey LAB
- ITBench SRE
- τ³-Bench Banking
- τ²-Bench Telecom