This site lets you try Janus Pro, an open-source multimodal model from DeepSeek, directly in your browser. It can create images from text prompts and also analyse and describe images you give it.
The model is released under the MIT license, so developers are free to build on it. The site claims strong results against well-known image generators on some benchmarks. It is free to use, making it an easy way for students and hobbyists in Pakistan to experiment with open-source image AI.
