Commits · 4b0b516544d5f4896acaa574aae18cba798a2be8 · chenpangpang / ComfyUI

09 Jun, 2023 1 commit
- Fix bug when embedding gets ignored because of mismatched size. · 23cf8ca7
  comfyanonymous authored Jun 08, 2023
  
  23cf8ca7
06 Jun, 2023 2 commits
- Small refactor. · 0e425603
  comfyanonymous authored Jun 06, 2023
  
  0e425603
- Refactor previews into one command line argument. · a3a713b6
  comfyanonymous authored Jun 06, 2023
```
Clean up a few things.
```
  a3a713b6
05 Jun, 2023 4 commits
- preview method autodetection · 3e17971a
  space-nuko authored Jun 05, 2023
  
  3e17971a
- Add latent2rgb preview · d5a28fad
  space-nuko authored Jun 05, 2023
  
  d5a28fad
- Make previews into cli option · 48f7ec75
  space-nuko authored Jun 05, 2023
  
  48f7ec75
- Preview sampled images with TAESD · b4f434ee
  space-nuko authored May 30, 2023
  
  b4f434ee
04 Jun, 2023 1 commit
- Some comments to say what the vram state options mean. · fed0a4dd
  comfyanonymous authored Jun 04, 2023
  
  fed0a4dd
03 Jun, 2023 2 commits
- Cleanups and fixes for model_management.py · 0a5fefd6
  comfyanonymous authored Jun 03, 2023
```
Hopefully fix regression on MPS and CPU.
```
  0a5fefd6
- Implement global average pooling for controlnet. · 700491d8
  comfyanonymous authored Jun 03, 2023
  
  700491d8
02 Jun, 2023 3 commits
- Refactor and improve model_management code related to free memory. · 67892b5a
  comfyanonymous authored Jun 02, 2023
  
  67892b5a
- More accurate total · 499641eb
  space-nuko authored Jun 02, 2023
  
  499641eb
- System stats endpoint · b5dd15c6
  space-nuko authored Jun 01, 2023
  
  b5dd15c6
01 Jun, 2023 2 commits
- Tweak lowvram model memory so it's closer to what it was before. · 5c38958e
  comfyanonymous authored Jun 01, 2023
  
  5c38958e
- Empty cache on mps. · 94680732
  comfyanonymous authored Jun 01, 2023
  
  94680732
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
29 May, 2023 1 commit
- Add route to get safetensors metadata: · b9818eb9
  comfyanonymous authored May 29, 2023
```
/view_metadata/loras?filename=lora.safetensors
```
  b9818eb9
28 May, 2023 2 commits
- Support VAEs in diffusers format. · a5328888
  comfyanonymous authored May 28, 2023
  
  a5328888
- Refactor diffusers model convert code to be able to reuse it. · 0fc483dc
  comfyanonymous authored May 28, 2023
  
  0fc483dc
25 May, 2023 4 commits
- Remove einops. · eb4bd771
  comfyanonymous authored May 25, 2023
  
  eb4bd771
- Do operations in same order as the one it replaces. · 87ab25fa
  comfyanonymous authored May 25, 2023
  
  87ab25fa
- Support old pytorch versions that don't have weights_only. · e1278fa9
  comfyanonymous authored May 25, 2023
  
  e1278fa9
- vecorized bislerp · 8b4b0c31
  BlenderNeko authored May 25, 2023
  
  8b4b0c31
23 May, 2023 2 commits
- Various improvements to bislerp. · b8ccbec6
  comfyanonymous authored May 23, 2023
  
  b8ccbec6
- Add experimental bislerp algorithm for latent upscaling. · 34887b88
  comfyanonymous authored May 23, 2023
```
It's like bilinear but with slerp.
```
  34887b88
22 May, 2023 1 commit
- Auto transpose images from exif data. · 6cc45057
  comfyanonymous authored May 22, 2023
  
  6cc45057
21 May, 2023 2 commits
- sample_dpmpp_2m_sde no longer crashes when step == 1. · dc198650
  comfyanonymous authored May 21, 2023
  
  dc198650
- Add DPM-Solver++(2M) SDE and exponential scheduler. · 069657fb
  comfyanonymous authored May 21, 2023
```
exponential scheduler is the one recommended with this sampler.
```
  069657fb
20 May, 2023 2 commits
- Make scaled_dot_product switch to sliced attention on OOM. · b8636a44
  comfyanonymous authored May 20, 2023
  
  b8636a44
- Simplify and improve some vae attention code. · 797c4e8d
  comfyanonymous authored May 20, 2023
  
  797c4e8d
15 May, 2023 1 commit
- Switch default scheduler to normal. · ef815ba1
  comfyanonymous authored May 15, 2023
  
  ef815ba1
13 May, 2023 2 commits

Print the torch device that is used on startup. · 3a1f4776
comfyanonymous authored May 13, 2023

3a1f4776

Make nodes map over input lists (#579) · 1201d2ea

BlenderNeko authored May 13, 2023

* allow nodes to map over lists

* make work with IS_CHANGED and VALIDATE_INPUTS

* give list outputs distinct socket shape

* add rebatch node

* add batch index logic

* add repeat latent batch

* deal with noise mask edge cases in latentfrombatch

1201d2ea

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
10 May, 2023 1 commit
- Auto batching improvements. · f7c0f75d
  comfyanonymous authored May 10, 2023
```
Try batching when cond sizes don't match with smart padding.
```
  f7c0f75d
09 May, 2023 1 commit
- Not needed anymore because sampling works with any latent size. · 314e526c
  comfyanonymous authored May 09, 2023
  
  314e526c
08 May, 2023 1 commit
- Make t2i adapter work with any latent resolution. · c6e34963
  comfyanonymous authored May 08, 2023
  
  c6e34963
06 May, 2023 1 commit
- Make maximum_batch_area take into account python2.0 attention function. · 6fc49176
  comfyanonymous authored May 06, 2023
```
More conservative xformers maximum_batch_area.
```
  6fc49176