Through our experiments, we enjoy challenging models with real tasks. Recently, we tested three models which are Qwen Image Edit Fast powered by Pruna AI, Nano Banana, and Flux Kontext Pro to handle a variety of real-world image editing tasks: from background swaps and object removal to style transformation.
Qwen Image Edit Fast (Pruna AI): Qwen-Image-Edit is the image editing version of Qwen-Image. It successfully extends Qwen-Image’s unique text rendering capabilities to editing tasks, enabling precise text modifications. The model is powered by Pruna AI.
Nano Banana: Gemini 2.5 Flash Image (aka nano-banana) is Google’s latest state-of-the-art image generation and editing model, designed to deliver higher-quality visuals and greater creative control. Building on the speed and efficiency of Gemini 2.0 Flash, this update introduces advanced features such as blending multiple images into one, maintaining character consistency for storytelling, applying precise edits through natural language, and leveraging Gemini’s world knowledge for richer results.
Flux Kontext Pro: Flux Kontext Pro, developed by Black Forest Labs, is a suite of generative flow-matching models built for both image creation and editing. It’s designed to grasp the context of images and text prompts, allowing users to generate or refine visuals with simple instructions while preserving consistency and high quality.
SeedEdit 3.0: It is the latest image editing model paired with Seedream 3.0, delivering major improvements in instruction following and content preservation on real images. It introduces a refined data curation pipeline with meta-info embedding to effectively scale editing data and connect VLMs with diffusion models more closely. A new joint learning approach combines diffusion loss with reward-based optimization, enabling higher-quality edits.
Each model brings something unique to the table — now let’s take a closer look at how they line up:
1. Prompt: Remove the blush from the face
2. Prompt: Add purple lipstick to the lips
3. Prompt: Remove hair from the woman
4. Prompt: Add afro hairstyle to the woman
5. Prompt: Remove the hat and sunglasses from the subject
6. Prompt: Change the subject’s clothing into traditional Japanese attire
7. Prompt: Remove the fog
8. Prompt: Add Spinosaurus to the background
9. Prompt: Change the direction of the man’s head so the camera captures it from the exact side profile
10. Prompt: Change the camera angle so it faces directly towards the dog
11. Prompt: Change the wall with a brick wall
12. Prompt: Write ‘Happy New Year’ on the blackboard
13. Prompt: Change the flower pot with a canvas
14. Prompt: Add an armchair and a coffee table next to the lamp
15. Prompt: Remove the headphones and cables
16. Prompt: Make the man look happy
17. Prompt: Make the man look sad
18. Prompt: Make the man strong and muscular
19. Prompt: Make the dog slim
20. Prompt: Change the text ‘STOP’ to ‘DUR’ on the sign, keeping the same style and deformation
21. Prompt: Convert the image into a 3D animated style
22. Prompt: Turn the hummingbird into a polished glass bird, transform the flower into a delicate crystal sculpture
23. Prompt: Replace all the bullets with shimmering, multi-colored butterflies
24. Prompt: Transform the scene into a peaceful winter landscape. Cover the ground and trees with a fresh layer of snow. Add gentle snowflakes falling from the sky, with a soft white overcast atmosphere
25. Prompt: Show this man eating breakfast at a table
26. Prompt: Replace the Taj Mahal with a detailed medieval castle, ensuring all reflections in water, glass, or surrounding surfaces match the new structure accurately. Keep lighting, shadows, and perspective consistent with the original scene for full realism
Here are the run times and prices per run:
| Flux Kontext Pro | Qwen Image Edit Fast | Nano Banana | SeedEdit 3.0 | |
| Run Time | ~10 seconds | ~8 seconds | ~7 seconds | ~14 seconds |
| Price | 0.04$ | ~0.03$ | 0.039$ | 0.03$ |
After going through these 25 prompts, it is clear that each model has its own strengths. Qwen Image Edit Fast stands out for precise text edits, Nano Banana offers creative flexibility, Flux Kontext Pro delivers strong context handling, and SeedEdit 3.0 focuses on faithful instruction following and content preservation. Together, they highlight how quickly image editing models are evolving.
Now it is your turn to experiment. Try the prompts, compare the results, and see which model fits your needs best. Share your experiences and ideas with the community so we can all learn more about what these tools can do.
Wiro AI, Machine Learning Team
