Commits · 3baded9892a6ac02f57caaf68053791ec0e14c5a · chenpangpang / ComfyUI

28 Apr, 2023 1 commit
- Basic torch_directml support. Use --directml to use it. · 3baded98
  comfyanonymous authored Apr 28, 2023
  
  3baded98
27 Apr, 2023 1 commit
- Add callback to sampler function. · 5a971cec
  comfyanonymous authored Apr 27, 2023
```
Callback format is: callback(step, x0, x)
```
  5a971cec
25 Apr, 2023 3 commits
- Some fixes to the batch masks PR. · aa57136d
  comfyanonymous authored Apr 25, 2023
  
  aa57136d
- Refactor more code to sample.py · c50208a7
  comfyanonymous authored Apr 24, 2023
  
  c50208a7
- This is cleaner this way. · 7983b3a9
  comfyanonymous authored Apr 24, 2023
  
  7983b3a9
24 Apr, 2023 2 commits
- gligen tuple · 0b07b2cc
  BlenderNeko authored Apr 24, 2023
  
  0b07b2cc
- made sample functions more explicit · d9b1595f
  BlenderNeko authored Apr 24, 2023
  
  d9b1595f
23 Apr, 2023 3 commits
- add docstrings · 58185397
  BlenderNeko authored Apr 23, 2023
  
  58185397
- refactor/split various bits of code for sampling · 2a09e2aa
  BlenderNeko authored Apr 23, 2023
  
  2a09e2aa
- Implement Linear hypernetworks. · 5282f564
  comfyanonymous authored Apr 23, 2023
```
Add a HypernetworkLoader node to use hypernetworks.
```
  5282f564
22 Apr, 2023 1 commit
- This makes pytorch2.0 attention perform a bit faster. · 6908f9c9
  comfyanonymous authored Apr 22, 2023
  
  6908f9c9
21 Apr, 2023 1 commit
- Remove some useless code. · 907010e0
  comfyanonymous authored Apr 20, 2023
  
  907010e0
20 Apr, 2023 1 commit
- Don't pass adm to model when it doesn't support it. · 96b57a9a
  comfyanonymous authored Apr 19, 2023
  
  96b57a9a
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 4 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
- fixed improper padding · d0b1b6c6
  BlenderNeko authored Apr 15, 2023
  
  d0b1b6c6
- Move code to empty gpu cache to model_management.py · deb2b93e
  comfyanonymous authored Apr 15, 2023
  
  deb2b93e
14 Apr, 2023 4 commits
- Safely load pickled embeds that don't load with weights_only=True. · 04d9bc13
  comfyanonymous authored Apr 14, 2023
  
  04d9bc13
- ensure backwards compat with optional args · da115bd7
  BlenderNeko authored Apr 14, 2023
  
  da115bd7
- align behavior with old tokenize function · 752f7a16
  BlenderNeko authored Apr 14, 2023
  
  752f7a16
- Don't stop workflow if loading embedding fails. · 334aab05
  comfyanonymous authored Apr 14, 2023
  
  334aab05
13 Apr, 2023 2 commits
- split tokenizer from encoder · 73175cf5
  BlenderNeko authored Apr 13, 2023
  
  73175cf5
- add unique ID per word/embedding for tokenizer · 8489cba1
  BlenderNeko authored Apr 13, 2023
  
  8489cba1
09 Apr, 2023 2 commits
- Fix for new transformers version. · 92eca60e
  comfyanonymous authored Apr 09, 2023
  
  92eca60e
- Print xformers version and warning about 0.0.18 · 1e1875f6
  comfyanonymous authored Apr 09, 2023
  
  1e1875f6
07 Apr, 2023 5 commits
- Clarify what --windows-standalone-build does. · 7e254d2f
  comfyanonymous authored Apr 07, 2023
  
  7e254d2f
- Cleanup. · 44fea050
  comfyanonymous authored Apr 07, 2023
  
  44fea050
- Fix loading SD1.5 diffusers checkpoint. · 58ed0f2d
  comfyanonymous authored Apr 07, 2023
  
  58ed0f2d
- Add a --force-fp32 argument to force fp32 for debugging. · 64557d67
  comfyanonymous authored Apr 07, 2023
  
  64557d67
- Small refactor. · bceccca0
  comfyanonymous authored Apr 06, 2023
  
  bceccca0
06 Apr, 2023 8 commits
- fixes lack of support for multi configs · 28fff5d1
  EllangoK authored Apr 06, 2023
```
also adds some metavars to argarse
```
  28fff5d1
- Rename the cors parameter to something more verbose. · f84f2508
  comfyanonymous authored Apr 06, 2023
  
  f84f2508
- makes cors a cli parameter · 48efae16
  EllangoK authored Apr 06, 2023
  
  48efae16
- set listen flag to listen on all if specifed · 01c1fc66
  EllangoK authored Apr 06, 2023
  
  01c1fc66
- Fix auto lowvram detection on CUDA · 3e2608e1
  藍+85CD authored Apr 06, 2023
  
  3e2608e1
- diffusers loader · 60127a83
  sALTaccount authored Apr 05, 2023
  
  60127a83
- Use separate variables instead of `vram_state` · 7cb924f6
  藍+85CD authored Apr 06, 2023
  
  7cb924f6
- Import intel_extension_for_pytorch as ipex · 84b9c0ac
  藍+85CD authored Apr 06, 2023
  
  84b9c0ac