Scene Text Editing on English Benchmark
88.57Sequence AccuracySelf-Prompting Diffusion Transformer
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Self-Prompting Diffusion Transformer2026.05 | 88.57 | 95.68 | 7.62 | 0.0365 | |
| FireRed-Image-Edit-1.12026.05 | 73.75 | 81.25 | 21.8 | 0.172 | |
| Qwen-Image-Edit-25112026.05 | 70.82 | 78.93 | 23.95 | 0.1874 | |
| Qwen-Image-Edit-25092026.05 | 69.02 | 74.22 | 24.2 | 0.1985 | |
| Longcat-Image-Edit2026.05 | 67.25 | 72.1 | 25.6 | 0.205 |