The way multi image mode works for FLUX Kontext (like here: https://
replicate.com/flux-kontext-a
pps/multi-image-kontext
…), essentially you just need to combine two images into a single image and send that to Kontext and it works out the rest. You can concatenate them side by side, above and below, or even something
@fofrai
-
FLUX Kontext Multi-Image Mode: Combining Images for AI Processing
By
–
-
AI Image Editing: Modern Camera Color Restoration Prompt
By
–
Try a prompt like this one: > Edit this image to be full color, as taken on a modern camera
-
Kontext Interprets Midjourney Generations with Advanced Prompt Engineering
By
–
I love how Kontext can interpret an old Midjourney generation > Make it a beautiful and epic and detailed but accurate photo, she is wearing large dark glass eyes
-
Kontext Image Expansion Adjusts Character Spacing
By
–
Notice how Kontext has expanded the image, but it's also put the two characters closer together.
-
Kontext: Change Image Aspect Ratio with AI Prompts
By
–
Here's a fun use of Kontext – you can use it to change the aspect ratio of an input. Change the aspect ratio from "match_input_image" to anything different. Prompt:
> make the image taller Probably pretty good for making iPhone wallpapers out of anything. -
Flux Kontext Launches Text-Removal AI Tool on Replicate
By
–
We made text-removal a thing https://
replicate.com/flux-kontext-a
pps/text-removal
… (thanks @lucataco93 ) -
Flux Kontext Launches Text-Removal Tool on Replicate
By
–
We made text-removal a thing https://
replicate.com/flux-kontext-a
pps/text-removal
… (thanks @lucataco93 ) -
Flux AI Enables Realistic Haircut Style Transfer Applications
By
–
nice hair https://
replicate.com/flux-kontext-a
pps/change-haircut
… -
AI System Demonstrates Strong Performance Under Varied Conditions
By
–
It's coped pretty well with everything I've thrown at it
-
Flux Kontext Portrait Series Model for Faster Cheaper Image Generation
By
–
We've had so much fun making these https://
replicate.com/flux-kontext-a
pps
… These are outputs from the "portrait-series" model, it's like my old consistent character model, but it's cheaper and faster, and also, wow. (Top left is the original)