E-Commerce Bench: leaderboard
Metric: Final Assets (¥k, mean of 5 episodes). Source: github.com. 18 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.6 Sol (Max) | 1431 |
| 2 | GPT-5.5 | 702 |
| 3 | Claude Opus 4.8 (Max) | 498 |
| 4 | GLM-5.2 (High) | 301 |
| 5 | Kimi K3 | 265 |
| 6 | Claude Opus 4.7 (Max) | 259 |
| 7 | Claude Opus 4.6 (Max) | 258 |
| 8 | GLM-5.1 | 226 |
| 9 | Gemini 3.5 Flash | 190 |
| 10 | Qwen 3.7 Max | 165 |
| 11 | Gemini 3.1 Pro (Preview) | 130 |
| 12 | GLM-5.2 (Max) | 115 |
| 13 | Kimi K2.6 | 70 |
| 14 | Qwen 3.6 Plus | 47 |
| 15 | Qwen 3.5 Plus | 1.1 |
Interactive version: theaggregate.ai/benchmark?slug=e-commerce-bench · How It Works · Data refreshed daily, snapshot 2026-09-05.