TinyGrafixBench - Bar Chart: leaderboard
TinyGrafixBench mIoU (%) on bar chart edits (adding, sorting, removing and recoloring bars), scored pixel by pixel against the single correct image; higher is better (PaintBench paper, NYU, 2026).
Metric: mIoU (%). Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 11 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Nano-Banana-2 | 38.9 | |
| 2 | GPT-Image-2 | 34.8 | |
| 3 | Nano Banana | 11.5 | |
| 4 | Qwen-Image-Edit-2511 | 4.8 | |
| 5 | LongCat-Image-Edit | 2.9 | |
| 6 | FLUX.2-klein-9B | 2.9 | |
| 7 | BAGEL-7B-MoT | 2.7 | #5515 |
| 8 | FLUX.2-dev | 2.3 | |
| 9 | FLUX.1-Kontext-dev | 1.3 | |
| 10 | HunyuanImage-3.0 | 0.2 |
Interactive version: theaggregate.ai/benchmark?slug=tinygrafixbench-bar-chart · How It Works · Data refreshed daily, snapshot 2026-11-12.