Skip to content

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

fd271de
Added "system_prompt" input to Gemini nodes (#11177)
Alexander Piskun committed 11 months ago
c3c6313
chore: replace imports of deprecated V1 classes (#11127)
Alexander Piskun committed 11 months ago
85c4b4a
Update workflow templates to v0.7.51 (#11150)
ComfyUI Wiki committed 11 months ago
058f084
ec7f651
Properly load the newbie diffusion model. (#11172)
comfyanonymous committed 11 months ago
56fa7db
Fix on-load VRAM OOM (#11144)
rattus committed 11 months ago
4086acf
Fix EmptyAudio node input types (#11149)
Jukka Seppänen committed 11 months ago
7ac7d69
marked all Pika API nodes a deprecated (#11146)
Alexander Piskun committed 11 months ago
76f18e9
Set OCL_SET_SVM_SIZE on AMD. (#11139)
comfyanonymous committed 11 months ago
d7a0aef
[V3] convert nodes_mask.py to V3 schema (#10669)
Alexander Piskun committed 11 months ago
913f86b
convert nodes_freelunch.py to the V3 schema (#10904)
Alexander Piskun committed 11 months ago
117bf3f
Fix regression. (#11137)
comfyanonymous committed 11 months ago
ae676ed
Kandinsky5 model support (#10988)
Jukka Seppänen committed 11 months ago
fd10932
Fix some custom nodes. (#11134)
comfyanonymous committed 11 months ago
092ee8a
Context windows fixes and features (#10975)
Jukka Seppänen committed 11 months ago
79d17ba
6fd463a
Forgot to put this in README. (#11112)
comfyanonymous committed 11 months ago
35fa091
[API Nodes]: fixes and refactor (#11104)
Alexander Piskun committed 11 months ago
3c84562
sd: bump HY1.5 VAE estimate (#11107)
rattus committed 11 months ago
9bc893c
sd: revise hy VAE VRAM (#11105)
rattus committed 11 months ago
f4bdf5f
Qwen Image Lora training fix from #11090 (#11094)
comfyanonymous committed 11 months ago
ecdc869