Image Editing
Discover and compare the best AI models for image editing and generation. Note: This is my personal informal leaderboard. Models are ranked by the completion rate of a series of diverse prompts designed to thoroughly assess performance.
| Rank | Company | Model | Score |
|---|---|---|---|
OpenAI | GPT Image 2.0 | 89 | |
Google | Nano Banana 2 (Gemini 3.1 Flash Image) | 88 | |
Google | 86 | ||
4 | OpenAI | 83 | |
5 | ByteDance Seed | 82 | |
6 | ByteDance Seed | 80 | |
6 | Alibaba | 80 | |
8 | Baidu | ERNIE Image | 77 |
8 | Black Forest Labs | 77 | |
10 | Google | 75 | |
11 | Alibaba | 71 | |
12 | OpenAI | GPT-4o | 69 |
13 | Black Forest Labs | 67 | |
13 | ByteDance Seed | SeedEdit 3.0 | 67 |
15 | Black Forest Labs | 66 | |
16 | Meituan | LongCat-Image-Edit | 65 |
17 | Black Forest Labs | 63 | |
18 | HiDream | 62 | |
19 | Black Forest Labs | 56 | |
20 | Google | Gemini 2.0 Flash Preview | 55 |
21 | Bytedance | 43 | |
21 | VectorSpaceLab | 43 | |
23 | StepFun | Step1X-Edit | 42 |
Full tutorial & review videos
Watch the videos below for comprehensive comparisons and detailed installation guides for select image editing models.
Methodology
Models are ranked using a series of prompts involving diverse range of challenging tasks. This includes:
- Prompt adherence and understanding
- Human anatomy
- Generating text
- Diagrams and infographics
- World understanding
- Uncommon poses and expressions
- Spatial understanding
- NSFW capabilities
To prevent manipulation, the prompts are kept confidential and are regularly updated to increase difficulty as models improve. Here is a subset of prompts for your reference:
Make the characters fight according to the pose drawing. They are fighting in an arctic background with a blizzard
Turn the scene into night time
Make a fisheye overhead view of this scene
Turn all the girls into cats and boys into dogs
Make a model sheet of this character with a front, back, and side view
Translate all the text into Chinese
Restore and colorize this photo