MMShopBench - Identifier Match@3: leaderboard
MMShopBench (2026) is a real-log benchmark for multimodal, multi-turn shopping agents that must recommend catalog products meeting intents and requirements expressed through images and dialogue. Share of cases where any of the first three products is a human-verified target, in percent, after evidence-grounded verification; higher is better.
Metric: Identifier match at 3 (%; share of cases in which any of the first three recommended products is one of the human-verified target product identifiers; multimodal multi-turn shopping requests from real assistant logs, answered by a tool-using agent over a frozen offline catalog with text and region-aware image retrieval (at most eight tool steps), followed by evidence-grounded verification and selection of the candidate pool by the policy model). Source: arxiv.org. Saturation forecast: Around February 2027. 7 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 64 | #94 |
| 2 | Claude Opus 4.8 (Thinking) | 59.9 | #56 (Claude Opus 4.8) |
| 3 | Kimi K2.6 (Thinking) | 38.8 | #178 (Kimi K2.6) |
| 4 | MiniMax-M2.7 | 21.5 | #518 |
| 5 | Qwen 3.5 9B (Non-reasoning) | 19.4 | #803 (Qwen 3.5 9B) |
| 6 | Qwen 3.5 27B (Non-reasoning) | 11.4 | #359 (Qwen 3.5 27B) |
| 7 | Qwen 3.5 122B A10B (Non-reasoning) | 5.9 | #331 (Qwen 3.5 122B A10B) |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=mmshopbench-identifier-match-3 · How It Works · Data refreshed daily, snapshot 2026-11-13.