tau-Rec: leaderboard

Metric: pass^1 (0-1 scaled to %), the mean strict success rate per trial: a recommendation succeeds only when every required typed constraint holds against the catalog and every active policy is respected; 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: Around December 2026. 9 models tracked.

Top models

#ModelScore
1DeepSeek V4 Flash (Max)57.1
2DeepSeek V4 Flash (High)56
3GPT-5.4 (Medium)55.1
4DeepSeek V4 Flash (Non-reasoning)54.6
5Claude Sonnet 4.653.7
6GPT-5.4 (Non-reasoning)47.1
7GPT-5 Mini41.7
8Gemini 2.5 Flash27.5
9Qwen 3 32B27.1

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