MMGist - Diagram and OCR: leaderboard
MMGist (2026) is a curated multimodal benchmark: 7,262 items kept from 18 vision-language benchmarks after dropping items that can be answered without the image, items nearly all models solve, and items with faulty labels. This board is accuracy in percent on the Diagram and OCR items, drawn from AI2D and OCRBench; higher is better.
Metric: Accuracy (%) on the Diagram and OCR dimension (items kept from AI2D and OCRBench); MMGist: 7,262 curated items from 18 vision-language benchmarks after removing items answerable without the image, items nearly every model solves and items with faulty labels; eight samples per item at temperature 1.0, step-by-step answer in a boxed block, 16,384 output tokens, medium effort where configurable; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 27 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) (Medium) | 82.3 | #94 (Gemini 3.1 Pro (Preview)) |
| 2 | Gemini 3.1 Flash Lite (Medium) | 73.4 | #364 (Gemini 3.1 Flash Lite) |
| 3 | Qwen 3.6 Plus | 68.7 | #222 |
| 4 | Seed 2.0 Pro | 65.8 | #211 |
| 5 | Qwen 3.5 9B | 65.5 | #803 |
| 6 | Qwen 3.6 35B A3B | 65.4 | #462 |
| 7 | Gemma 4 31B | 63.8 | #394 |
| 8 | Seed 2.0 Lite | 63.2 | #289 |
| 9 | Seed 2.0 Mini | 63.2 | #508 |
| 10 | Gemma 4 26B A4B | 61.3 | #615 |
| 11 | Claude Sonnet 4.6 (Medium) | 61.3 | #161 (Claude Sonnet 4.6) |
| 12 | GPT-5 (Medium) | 58.6 | #167 (GPT-5) |
| 13 | Qwen 3.5 27B | 55.4 | #359 |
| 14 | GPT-5 Mini (Medium) | 54.4 | #348 (GPT-5 Mini) |
| 15 | Qwen 3.5 4B | 53.7 | #1175 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=mmgist-diagram-and-ocr · How It Works · Data refreshed daily, snapshot 2026-11-13.