Allow image generation to use reference image.
Venice has ability to analyze an uploaded image, and you have multiple image generation models, but they are only text to image, not text + reference image to image.
All of the venice text to image models have difficulty positioning bodies precisely (such as in the schoolgirl pin wrestling position the wrestler is always in squat position with knees in air instead of against the mat and no level of prompting seems able to get the knees pressed down. A reference image would be very useful in such scenarios.
Log in to comment and vote
Comments18
Red Spray
Jun 8, 2025
Seconded
Amethyst Frost
Aug 3, 2025
Also if we had this, we could get the ai to replicate specific human expressions easier because then we could just show a picture to it and then it can map that expression. Which is exactly what ChatGPT does already as well Gemini
Amethyst Frost
Aug 3, 2025
This is secretly one of those essential features that any intelligent image generator would have. You could even role out this feature with the in development Uncensored Model with Vision. We really just need the uncensored image generators to have vision. Kind of wild that an image generator doesn't have vision.
Blue Orbit
Apr 24, 2025
•Merged request
•4 votes
Ability to view images on websites and use them to improve results.
It would help with references and the model making sure its following a character design to a tee. Grok has this feature if you need something to use as a base.
An Anonymous User
Apr 23, 2025
•Merged request
•6 votes
Generate a cartoon from uploaded image
Wouldn’t it be amazing to upload a picture - for example- your cat- and get the AI to turn it into an accurate cartoon? This would be far superior to describing your cat in words using a text prompt. You could then name the character and ask the Ai to put it into different imaginary situations- Mr Tibbles goes skiing.
An Anonymous User
Feb 26, 2025
•Merged request
•5 votes
Convert an uploaded Image
Change background and other features in an uploaded Image.
Silver River
Apr 24, 2025
Often referred to as image to image.
Tan Crocodile
Mar 25, 2025
•Merged request
•3 votes
give the image generator a web search
The image generators do not do search web behind generating, entering specific prompts is not enough make one that web searches
An Anonymous User
Aug 8, 2025
•Merged request
•3 votes
Just one thing missing..
You’re by far my favourite AI system- from your image generation to your chatbots just wow… thank you seriously !
I have one suggestion though, we should be able to give an example image or two when generating images through any of the available ones- it would really help creators who make fanart of their favourite tv and movie characters- I’d pay extra too on top of the premium subscription if you guys ever made image to video too !
Thanks for reading :))
An Anonymous User
Aug 8, 2025
•Merged request
•2 votes
Reference image
Allow using reference images for image creation. In ChatGPT, it’s possible to upload an image and use it as a part of the prompt, that would be nice. Also, it would be nice to not have to switch between text and image, but have them in the same chat
Jack
Jul 30, 2025
Try out image editing for this:
https://venice.ai/faqs#how-do-i-edit-an-image-on-venice
Cyan Snake
Aug 19, 2025
I need this!
Rose Newt
Sep 17, 2025
Absolutely support this. I’ve been using and fine tuning chatGPT character faces and used them in image generation with a transparent contouring mask and it really helped generating consistant character faces and traits. Which is even more immersive in story telling experience.