EntCollabBench - Approval: leaderboard
Metric: Task success (%) on the 100 policy approval tasks (80 single-step and 20 multi-step), judged against deterministic policy adjudications, role-specialized LLM agents (11 enterprise roles behind permission-controlled ITSM, HR, CSM, Gitea and document services) delegate over HTTP, temperature 0, three-judge majority vote; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.6 | 70 |
| 2 | DeepSeek V4 Flash | 70 |
| 3 | DeepSeek V4 Pro | 68 |
| 4 | GPT-5 Mini | 62 |
| 5 | GPT-5.4 | 60 |
| 6 | Gemini 3.1 Pro (Preview) | 58 |
| 7 | Gemini 3.1 Flash Lite (Preview) | 46 |
| 8 | Qwen 3.5 122B A10B | 45 |
| 9 | Qwen 3.5 35B A3B | 43 |
| 10 | MiniMax-M2.7 | 37 |
| 11 | Qwen 3.5 9B | 23 |
| 12 | MiMo-V2-Flash | 7 |
Interactive version: theaggregate.ai/benchmark?slug=entcollabbench-approval · How It Works · Data refreshed daily, snapshot 2026-10-07.