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
Support llama hunyuan video text encoder in scaled fp8 format.
d6656b0 comfyanonymous committed 1 year ago
Adjust used dtypes for hunyuan video VAE and diffusion model.
39b1fc4 comfyanonymous committed 1 year ago
use String directly to set bg color for load 3d canvas (#6057)
cc550d5 Terry Jia committed 1 year ago
Fix case of ExecutionBlocker not handled correctly with INPUT_IS_LIST.
6d1a3f7 comfyanonymous committed 1 year ago
fix: prestartup script is not applied due to `extra_model_paths.yaml` and ensure custom paths are used during startup (#5872)
558b7d8 Dr.Lt.Data committed 1 year ago
add_model_folder_path: ensure unique paths by removing duplicates (#5998)
caf2074 Alexander Piskun committed 1 year ago
Set initial load devices for te and model to mps device on mac.
fd5dfb8 comfyanonymous committed 1 year ago