MMM-Bench - Level 3 (Business Stage): leaderboard
MMM-Bench (2026, Alibaba) zero-shot document classification at taxonomy level 3, Business Stage, for real enterprise documents classified using the paper’s hierarchical category prompt. Macro-F1 in percent; higher is better.
Metric: Macro-F1 (%) at taxonomy level 3 (Business Stage), zero-shot hierarchical classification of the test split (20 percent of 5,990 real Alibaba business documents in 12 domains, mostly Chinese) with a document prompt containing OCR text, image-page placeholders and optional filename; actual image-tensor input is described only for multimodal implementations against a flattened five-level taxonomy, temperature 0.1; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 15 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 75.91 | #99 |
| 2 | Claude Opus 4.6 | 73.23 | #113 |
| 3 | Claude Opus 4.5 | 69.74 | #160 |
| 4 | Kimi K2.5 | 64.59 | #266 |
| 5 | Qwen 3.6 Plus | 64.51 | #238 |
| 6 | GLM-5 | 63.95 | #269 |
| 7 | GPT-5.4 | 62 | #135 |
| 8 | GPT-5.2 | 60.59 | #185 |
| 9 | MiniMax-M2.7 | 58.54 | #539 |
| 10 | Qwen 3.5 27B | 56.88 | #375 |
| 11 | Qwen 3 VL 235B A22B | 51.99 | #485 |
| 12 | Qwen 3.5 35B A3B | 49.53 | #492 |
| 13 | Qwen 3.5 9B | 41.02 | #828 |
| 14 | Qwen 3 VL 8B | 11.26 | #917 |
| 15 | Qwen 2.5 VL 7B | 6.39 | #1862 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=mmm-bench-level-3-business-stage · How It Works · Data refreshed daily, snapshot 2026-11-21.