Highlights
10 imagesBest models, by
Best Image Editing Models by Price
1 model without pricing omitted.
Best Image Editing Models by Speed
8 models waiting for enough speed data.
Best AI Models for Adding & Editing Objects
| # | Model | Elo |
|---|---|---|
| 1 | 1263 | |
| 2 | 1263 | |
| 3 | 1253 | |
| 4 | 1251 | |
| 5 | 1242 | |
| 6 | 1234 | |
| 7 | 1228 | |
| 8 | 1227 | |
| 9 | 1224 | |
| 10 | 1210 |
| 11 | 1209 | |
| 12 | 1205 | |
| 13 | 1196 | |
| 14 | 1195 | |
| 15 | 1193 | |
| 16 | 1189 | |
| 17 | 1187 | |
| 18 | 1176 | |
| 19 | 1173 | |
| 20 | 1165 | |
| 21 | 1153 | |
| 22 | 1152 | |
| 23 | 1151 | |
| 24 | 1144 | |
| 25 | 1143 | |
| 26 | 1136 | |
| 27 | 1134 | |
| 28 | 1098 | |
| 29 | 1097 |
Alibaba’s Wan 2.6 and Google’s Nano Banana Pro lead the category with an identical 1263 Elo, though Wan 2.6 offers a superior 69.5% win rate at less than half the price of its competitor. Notably, the budget-friendly GPT Image 1.5 outperforms more expensive FLUX.2 and Qwen iterations in editing tasks, securing a top-three position despite its low $0.009 per image cost.
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.