ShoppingMMLU: leaderboard
Online-shopping benchmark covering concept understanding, shopping knowledge reasoning, user-behavior alignment, and multilingual shopping abilities across 57 Amazon-derived tasks.
Metric: Average Score (%). Source: github.com. Source: benchmarklist.com. Saturation forecast: Estimated already saturated. 27 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude 3 Sonnet | 72.58 | #775 |
| 2 | Llama 3 70B Instruct | 68.55 | #634 |
| 3 | Claude 2 | 67.43 | #761 |
| 4 | Qwen 1.5 72B | 67.39 | #753 |
| 5 | ChatGPT | 65.3 | #900 |
| 6 | Llama 3 70B | 61.97 | #631 |
| 7 | Qwen 1.5 14B | 59.57 | #1096 |
| 8 | Llama 3 8B Instruct | 58.09 | #1129 |
| 9 | Mixtral 8x7B | 53.44 | #1143 |
| 10 | vicuna-13B-v1.5 | 52.93 | #1392 |
| 11 | Qwen 1.5 7B | 52.8 | #1327 |
| 12 | Llama 2 70B | 52 | #1201 |
| 13 | Llama 3 8B | 50.74 | #1242 |
| 14 | Qwen 1.5 4B | 50.57 | #1421 |
| 15 | Mistral 7B Instruct | 48.48 | #1363 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=shoppingmmlu · How It Works · Data refreshed daily, snapshot 2026-10-20.