TinyGrafixBench - Heatmap: leaderboard
Metric: mIoU (%). Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Nano-Banana-2 | 20.2 |
| 2 | GPT-Image-2 | 15.5 |
| 3 | LongCat-Image-Edit | 7.9 |
| 4 | Qwen-Image-Edit-2511 | 4.9 |
| 5 | FLUX.2-klein-9B | 4.8 |
| 6 | Nano Banana | 4 |
| 7 | FLUX.1-Kontext-dev | 3.5 |
| 8 | FLUX.2-dev | 3.2 |
| 9 | BAGEL-7B-MoT | 2 |
| 10 | HunyuanImage-3.0 | 0.9 |
Interactive version: theaggregate.ai/benchmark?slug=tinygrafixbench-heatmap · How It Works · Data refreshed daily, snapshot 2026-09-26.