Highlights
10 images
Best models, by
Best Image Editing Models by Price
1 model without pricing omitted.
Best Image Editing Models by Speed
Best AI Models for Lighting, Shadows & Materials
| # | Model | Elo |
|---|---|---|
| 1 | 1274 | |
| 2 | 1257 | |
| 3 | 1255 | |
| 4 | 1250 | |
| 5 | 1246 | |
| 6 | 1238 | |
| 7 | 1238 | |
| 8 | 1237 | |
| 9 | 1234 | |
| 10 | 1231 |
| 11 | 1230 | |
| 12 | 1229 | |
| 13 | 1225 | |
| 14 | 1224 | |
| 15 | 1217 | |
| 16 | 1217 | |
| 17 | 1210 | |
| 18 | 1207 | |
| 19 | 1189 | |
| 20 | 1174 | |
| 21 | 1172 | |
| 22 | 1167 | |
| 23 | 1159 | |
| 24 | 1146 | |
| 25 | 1133 | |
| 26 | 1133 | |
| 27 | 1107 | |
| 28 | 1103 | |
| 29 | 1080 |
Google dominates the top of the lighting and materials category, with Nano Banana Pro (Elo 1273) and Nano Banana 2 (Elo 1257) holding the highest rankings despite being significantly slower than the rest of the field. While GPT Image 2 achieves a category-high 85.7% win rate, it trails in Elo behind Google's models and the budget-friendly GPT Image 1.5 (Elo 1255).
Rivalries
Aggregate head-to-head across the arena
Highlighted challenges
Man and Car in California
Combines two references (a person and a car) into one believable driving scene. It checks whether the model keeps the man’s identity and the car’s appearance while placing them convincingly on the coastline with matched lighting and perspective.
Bald man challenge
A targeted edit test. Adding a realistic head of hair requires a believable hairline and texture while keeping the face, expression, and lighting unchanged. Common failures are wig-like or pasted-on hair and subtle changes to the underlying face.
Keep the arena honest
Cast your vote
Pick winners in blind matchups. Every vote nudges the Elo and shapes these rankings.
Cast Your VoteSuggest a prompt
Got an idea worth testing? Submit a prompt and watch the models battle it out.