Larry Wang
larrywang233
AI & ML interests
None yet
Recent Activity
new activity
about 1 month ago
NexaAI/OmniNeural-4B:Does the model run on CPU?
new activity
about 1 month ago
NexaAI/OmniNeural-4B:Is this supported on ollama/llama.cpp?
reacted
to
MonsterMMORPG's
post
with 😎
about 1 month ago
Wan 2.2, FLUX, FLUX Krea & Qwen Image Just got Upgraded: Ultimate Tutorial for Open Source SOTA Image & Video Gen Models - With easy to use SwarmUI with ComfyUI Backend : https://youtu.be/3BFDcO2Ysu4
Tutorial Video : https://youtu.be/3BFDcO2Ysu4
Wan 2.2, Qwen Image, FLUX, FLUX Krea, all these models are the SOTA open-source models and in this master tutorial I will show you how to use these models in the easiest, most performant, and most accurate way. After doing almost one week of research, I have determined the very best presets and prepared this tutorial. With literally one click you will be able to install, download models, set presets, and use these amazing models. Wan 2.2 is currently the king of video generation models and now it is super fast with lightx2v Wan2.2-Lightning LoRAs. Moreover, Qwen Image is now ultra-fast with the recently released 8-step LoRA with almost no quality loss. Furthermore, I have updated FLUX and FLUX Krea presets to improve image generation quality. Finally, I have trained FLUX Krea with our existing DreamBooth and LoRA training workflows, analyzed and shared the results in this tutorial. As additional information, I have shown the upcoming Qwen Image editing/inpainting model and the Qwen Image training application I am developing.
Organizations
None yet