100% Open Image Model Lets You Place Text With JSON

In this video I look at MOSS Image. A 5B diffusion transformer, which has an Apache 2 license, allowing you to create images and do whatever you want with it. It also allows you to use JSON to layout the image Blog: https://openmoss.ai/MOSS-Image/ HF: https://huggingface.co/OpenMOSS-Team/MOSS-Image-SFT 2nd Channel: https://www.youtube.com/channel/UCUF7WtR4JpVbQqcQGrLeADw Twitter: https://x.com/Sam_Witteveen 🕵️ Interested in building LLM Agents? Fill out the form below Building LLM Agents Form: https://drp.li/dIMes 👨‍💻Github: https://github.com/samwit/llm-tutorials ⏱️Time Stamps: 00:00 Intro 00:29 What MOSS Image 01:58 Where to run it: Hugging Face, ComfyUI, SGLang, vLLM-Omni 02:11 General images 02:38 Lighting, styles and aspect ratios 03:21 Text rendering 04:01 Long-text benchmark scores 04:29 Where the text breaks down 04:50 JSON layout control 06:08 What it can't do 07:53 Demo 09:00 Editing the layout 10:07 Styles and inpainting 10:44 8-step LoRA 11:21 Custom LoRA