tau-Rec - Policy Compliance: leaderboard

Metric: Mean policy sub-score (0-1 scaled to %), programmatic compliance with the seven recommendation policies (recommend tool, watch history, availability, age restriction, sponsorship disclosure, transparency, single recommendation); the agent recommends one movie from a 153-title TMDB catalog to a GPT-5 mini user simulator through dialogue and catalog tools on 60 tasks (constraints revealed up front, only on asking, or never), 4 trials per task, agent temperature 0 where supported; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 9 models tracked.

Top models

#ModelScore
1GPT-5 Mini96.3
2GPT-5.4 (Medium)95.8
3DeepSeek V4 Flash (Max)94.2
4DeepSeek V4 Flash (High)93.7
5DeepSeek V4 Flash (Non-reasoning)92.1
6Gemini 2.5 Flash92
7GPT-5.4 (Non-reasoning)90.8
8Claude Sonnet 4.689.6
9Qwen 3 32B75.6

Interactive version: theaggregate.ai/benchmark?slug=tau-rec-policy-compliance · How It Works · Data refreshed daily, snapshot 2026-09-29.