EnterpriseBench Sandbox (LangChain, CoT): leaderboard

Metric: Judged task accuracy (0-1; 500 enterprise tasks, final execution graded with the Prometheus-2 rubric by GPT-4; agent built in LangChain with two few-shot examples per call, chain-of-thought planning). Source: arxiv.org. Saturation forecast: Estimated already saturated. 5 models tracked.

Top models

#ModelScore
1O1 Mini0.28
2GPT-4o0.27
3Claude 3.5 Sonnet (20240620)0.27
4Llama 3.3 70B Instruct0.22
5Llama 3.1 8B Instruct0.06

Interactive version: theaggregate.ai/benchmark?slug=enterprisebench-sandbox-langchain-cot · How It Works · Data refreshed daily, snapshot 2026-09-26.