EnterpriseOps-Gym-AA
Enterprise tool-use benchmark of agents operating inside enterprise IT-service style environments.
Actively evaluated by Artificial Analysis on current frontier models; no saturation evidence recorded.
Performance Timeline
Longitudinal progression of model scores against human baselines.Performance & Historical Trajectory
Empirical score progression across model release dates and evaluation rounds.
| Model | Score | Date | Source Type | Provenance |
|---|---|---|---|---|
| DeepSeek V4 Flash Vision | 47.1% | 2026-11-15 | independent | Source ↗ |
| Inkling 2 Small | 46.8% | 2026-11-15 | independent | Source ↗ |
| Mistral Medium 3.5 | 33.7% | 2026-11-15 | independent | Source ↗ |
| Gemma 4 31B | 28.3% | 2026-11-15 | independent | Source ↗ |
| GLM 5.3 | 36.4% | 2026-11-15 | independent | Source ↗ |
| gpt-oss-120b | 25.5% | 2026-11-15 | independent | Source ↗ |
| Gemini 3.1 Pro | 42.2% | 2026-11-15 | independent | Source ↗ |
| Qwen3.8 Max (0902) | 47.6% | 2026-11-15 | independent | Source ↗ |
| Qwen3.7 Plus | 40.6% | 2026-11-15 | independent | Source ↗ |
| Qwen3.8 2.4T A95B | 47.4% | 2026-11-15 | independent | Source ↗ |
| Kimi K2.7 Code | 40.2% | 2026-11-15 | independent | Source ↗ |
| DeepSeek V4-Pro | 49.6% | 2026-11-15 | independent | Source ↗ |
| Gemini 3.5 Flash-Lite | 42.3% | 2026-11-15 | independent | Source ↗ |
| Inkling Small | 42.7% | 2026-11-15 | independent | Source ↗ |
| Nemotron 3 Ultra | 28.9% | 2026-11-15 | independent | Source ↗ |
| Muse Glimmer | 34.7% | 2026-11-15 | independent | Source ↗ |
| Qwen3.8 27B | 44.2% | 2026-11-15 | independent | Source ↗ |
| Nemotron 3.5 Lightning | 18.4% | 2026-11-15 | independent | Source ↗ |
| MiniMax M3 | 32.1% | 2026-11-15 | independent | Source ↗ |
| GLM 5.3 Flash | 33.2% | 2026-11-15 | independent | Source ↗ |
| Qwen3.5 397B A17B | 30.9% | 2026-11-15 | independent | Source ↗ |
| Inkling | 38% | 2026-11-15 | independent | Source ↗ |
| Step 5 | 47.2% | 2026-11-15 | independent | Source ↗ |
Human Baseline & Difficulty Horizon
Calibrated human reference points, specialist benchmarks, and ceiling thresholds.No peer-reviewed human baseline has been published for this benchmark.
Metric & Scoring Methodology
Verification protocols, aggregation formulas, and specialized metric variants.task success (%)Dataset & Compute Cost
Evaluation volume, public availability, API pricing, and local hardware requirements.$5 – $20 USD for full benchmark evaluation run on frontier APIs.
1x NVIDIA RTX 4090 (24GB) or A100 (40GB/80GB) via vLLM / SGLang
How to Run & Reproduce
Standardized evaluation protocols, CLI commands, and reproducible runner templates.lm_eval --model hf --model_args pretrained=<model_path> --tasks enterprise-ops-gym-aa --batch_size autoopencompass --datasets enterprise-ops-gym-aa --models <model_config># Standard API Evaluation Loop
from openai import OpenAI
client = OpenAI()
response = client.chat.completions.create(
model="gpt-4o",
messages=[{"role": "user", "content": prompt}],
temperature=0.0,
)When publishing results for EnterpriseOps-Gym-AA, always report the exact prompt template, few-shot exemplar ordering, sampling temperature (temperature=0), maximum reasoning budget tokens, and the precise timestamped model snapshot ID.
Contamination & Memorization Analysis
Audit of pretraining exposure risks, memorization vectors, and refresh policies.Static fixed snapshot
Public on web / HuggingFace