Commits · 05676942b73858d4cf2d0b6bc8680b52c5aa9f96 · chenpangpang / ComfyUI

24 Jun, 2023 1 commit
- Add some more transformer hooks and move tomesd to comfy_extras. · 05676942
  comfyanonymous authored Jun 23, 2023
```
Tomesd now uses q instead of x to decide which tokens to merge because
it seems to give better results.
```
  05676942
23 Jun, 2023 3 commits
- Move latent scale factor from VAE to model. · 8607c2d4
  comfyanonymous authored Jun 23, 2023
  
  8607c2d4
- Fix bug when yaml config has no clip params. · 30a38619
  comfyanonymous authored Jun 23, 2023
  
  30a38619
- Fix error with ClipVision loader node. · 9e37f4c7
  comfyanonymous authored Jun 23, 2023
  
  9e37f4c7
22 Jun, 2023 2 commits
- Don't merge weights when shapes don't match and print a warning. · 9f83b098
  comfyanonymous authored Jun 22, 2023
  
  9f83b098
- Support base SDXL and SDXL refiner models. · f87ec10a
  comfyanonymous authored Jun 22, 2023
```
Large refactor of the model detection and loading code.
```
  f87ec10a
20 Jun, 2023 3 commits
- Fix last commits causing an issue with the text encoder lora. · 51581dbf
  comfyanonymous authored Jun 20, 2023
  
  51581dbf
- Keep a set of model_keys for faster add_patches. · 8125b51a
  comfyanonymous authored Jun 20, 2023
  
  8125b51a
- Add a type of model patch useful for model merging. · 45beebd3
  comfyanonymous authored Jun 20, 2023
  
  45beebd3
19 Jun, 2023 1 commit
- Add a way to set patches that modify the attn2 output. · 8883cb0f
  comfyanonymous authored Jun 18, 2023
```
Change the transformer patches function format to be more future proof.
```
  8883cb0f
18 Jun, 2023 1 commit
- This is not needed anymore and causes issues with alphas_cumprod. · fb4bf7f5
  comfyanonymous authored Jun 18, 2023
  
  fb4bf7f5
15 Jun, 2023 1 commit
- Add a --gpu-only argument to keep and run everything on the GPU. · f7edcfd9
  comfyanonymous authored Jun 15, 2023
```
Make the CLIP model work on the GPU.
```
  f7edcfd9
14 Jun, 2023 1 commit
- Set model to fp16 before loading the state dict to lower ram bump. · 6b774589
  comfyanonymous authored Jun 14, 2023
  
  6b774589
13 Jun, 2023 3 commits
- sampler_cfg_function now uses a dict for the argument. · 388567f2
  comfyanonymous authored Jun 13, 2023
```
This means arguments can be added without issues.
```
  388567f2
- Turn on safe load for a few models. · ff9b22d7
  comfyanonymous authored Jun 13, 2023
  
  ff9b22d7
- Cleanup: Remove a bunch of useless files. · f0a2b81c
  comfyanonymous authored Jun 13, 2023
  
  f0a2b81c
12 Jun, 2023 2 commits
- Split the batch in VAEEncode if there's not enough memory. · f8c59310
  comfyanonymous authored Jun 12, 2023
  
  f8c59310
- Auto switch to tiled VAE encode if regular one runs out of memory. · c069fc07
  comfyanonymous authored Jun 11, 2023
  
  c069fc07
09 Jun, 2023 1 commit
- Simpler base model code. · de142eaa
  comfyanonymous authored Jun 09, 2023
  
  de142eaa
06 Jun, 2023 1 commit
- Small refactor. · 0e425603
  comfyanonymous authored Jun 06, 2023
  
  0e425603
03 Jun, 2023 1 commit
- Implement global average pooling for controlnet. · 700491d8
  comfyanonymous authored Jun 03, 2023
  
  700491d8
31 May, 2023 1 commit
- This is useless for inference. · 03da8a34
  comfyanonymous authored May 31, 2023
  
  03da8a34
30 May, 2023 1 commit
- Auto load model in lowvram if not enough memory. · eb448dd8
  comfyanonymous authored May 30, 2023
  
  eb448dd8
28 May, 2023 1 commit
- Support VAEs in diffusers format. · a5328888
  comfyanonymous authored May 28, 2023
  
  a5328888
12 May, 2023 2 commits
- comment out annoying print statement · 19c014f4
  BlenderNeko authored May 12, 2023
  
  19c014f4
- minor changes for tiled sampler · d9e088dd
  BlenderNeko authored May 12, 2023
  
  d9e088dd
04 May, 2023 1 commit
- Fix imports. · bae4fb4a
  comfyanonymous authored May 04, 2023
  
  bae4fb4a
03 May, 2023 4 commits
- Refactor. · fcf513e0
  comfyanonymous authored May 03, 2023
  
  fcf513e0
- remove unused import · 5eeecf3f
  pythongosssss authored May 03, 2023
  
  5eeecf3f
- use comfy progress bar · 8912623e
  pythongosssss authored May 03, 2023
  
  8912623e
- reduce duplication · 27df7410
  pythongosssss authored May 03, 2023
  
  27df7410
02 May, 2023 1 commit
- added progress to encode + upscale · 06ad35b4
  pythongosssss authored May 02, 2023
  
  06ad35b4
01 May, 2023 1 commit
- LoKR support. · 9c335a55
  comfyanonymous authored May 01, 2023
  
  9c335a55
24 Apr, 2023 1 commit
- Add progress to vae decode tiled · c8c9926e
  pythongosssss authored Apr 24, 2023
  
  c8c9926e
23 Apr, 2023 1 commit
- Implement Linear hypernetworks. · 5282f564
  comfyanonymous authored Apr 23, 2023
```
Add a HypernetworkLoader node to use hypernetworks.
```
  5282f564
19 Apr, 2023 1 commit
- Add support for GLIGEN textbox model. · 3696d169
  comfyanonymous authored Apr 19, 2023
  
  3696d169
17 Apr, 2023 1 commit
- Add a way for nodes to set a custom CFG function. · 884ea653
  comfyanonymous authored Apr 17, 2023
  
  884ea653
15 Apr, 2023 2 commits
- Fix model_management import so it doesn't get executed twice. · 73c3e11e
  comfyanonymous authored Apr 15, 2023
  
  73c3e11e
- Some refactoring: from_tokens -> encode_from_tokens · 81d1f00d
  comfyanonymous authored Apr 15, 2023
  
  81d1f00d
14 Apr, 2023 1 commit
- ensure backwards compat with optional args · da115bd7
  BlenderNeko authored Apr 14, 2023
  
  da115bd7