Diagram-MMU - Code Parsing: leaderboard
Metric: All Score (%). Source: vi-ocean.github.io. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Kimi K2.5 | 56.97 |
| 2 | Qwen 3 VL 235B A22B | 55.98 |
| 3 | Qwen 3.5 397B A17B | 52.72 |
| 4 | GPT-5.2 | 51.88 |
| 5 | Seed 2.0 Pro | 50.89 |
| 6 | Gemini 3.1 Pro (Preview) | 48.38 |
| 7 | InternVL3-38B | 41.13 |
Interactive version: theaggregate.ai/benchmark?slug=diagram-mmu-code-parsing · How It Works · Data refreshed daily, snapshot 2026-09-05.