Qwen Releases E-Commerce Bench: No Model Masters Year-Long Store Operation

Alibaba's Qwen team released E-Commerce Bench, a year-long simulation where agents run multiple online stores with 100,000 yuan starting capital. Among 18 frontier models, none achieved overall leadership and almost none turned a profit, with only two learning to negotiate prices.

2026-09-02 ~ 2026-09-03 · 4 related posts

• Qwen team's E-Commerce Bench runs 18 frontier models through a simulated year; no model dominates — dair_ai · 2026-09-02
• E-Commerce Bench: Evaluating LLM Agents on Long-Horizon Autonomous Business Operation — Qwen · 2026-09-02
• Qwen's E-CommerceBench makes 18 models run an online shop for a year — 千问大模型 · 2026-09-03
• Qwen Releases E-Commerce Bench: Agents Run Stores for 365 Days, Few Learn — Alibaba_Qwen · 2026-09-03