RealWorldShop: leaderboard

Metric: Session success rate (%): share of simulated shopping sessions that end in a grounded, actionable purchase decision (Kimi-K2.6 user simulator, at most 30 assistant turns); higher is better. Source: arxiv.org. Saturation forecast: Around December 2026. 7 models tracked.

Top models

#ModelScore
1GPT-571
2Kimi K2.667
3DeepSeek V3.256
4Qwen 3.5 27B55
5Qwen 3.5 35B A3B53
6GPT-4o38
7Gemini 2.5 Flash38

Interactive version: theaggregate.ai/benchmark?slug=realworldshop · How It Works · Data refreshed daily, snapshot 2026-10-02.