The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface. The fastest local inference engine in the world.
Commits on master
Fix RepeatLatentBatch not working on multi dim latents. (#9227)
05df2df comfyanonymous committed 1 year ago
Fixed an issue where qwenLora could not be loaded properly. (#9208)
4c3e57b flybirdxx committed 1 year ago
Fix default shift and any latent size for qwen image model. (#9186)
d044a24 comfyanonymous committed 1 year ago
Add Veo3 video generation node with audio support (#9110)
f69609b Christian Byrne committed 1 year ago
Add some warnings and prevent crash when cond devices don't match. (#9169)
84f9759 comfyanonymous committed 1 year ago
Various fixes for broken things from earlier PR. (#9168)
7991341 comfyanonymous committed 1 year ago
Lower cond vram use by casting at the same time as device transfer. (#9159)
182f90b comfyanonymous committed 1 year ago
Make sure all the conds are on the right device. (#9151)
5f582a9 comfyanonymous committed 1 year ago
Fix WanFirstLastFrameToVideo node when no clip vision. (#9134)
5ee381c comfyanonymous committed 1 year ago