comfyanonymous
|
4a77fcd6ab
|
Only shift text encoder to vram when CPU cores are under 8.
|
1 year ago |
|
comfyanonymous
|
3cd31d0e24
|
Lower CPU thread check for running the text encoder on the CPU vs GPU.
|
1 year ago |
|
comfyanonymous
|
2b13939044
|
Remove some useless code.
|
1 year ago |
|
comfyanonymous
|
95d796fc85
|
Faster VAE loading.
|
1 year ago |
|
comfyanonymous
|
4b957a0010
|
Initialize the unet directly on the target device.
|
1 year ago |
|
comfyanonymous
|
c910b4a01c
|
Remove unused code and torchdiffeq dependency.
|
1 year ago |
|
comfyanonymous
|
1141029a4a
|
Add --disable-metadata argument to disable saving metadata in files.
|
1 year ago |
|
comfyanonymous
|
68be24eead
|
Remove some prints.
|
1 year ago |
asagi4
|
1ea4d84691
|
Fix timestep ranges when batch_size > 1
|
1 year ago |
|
comfyanonymous
|
5379051d16
|
Fix diffusers VAE loading.
|
1 year ago |
|
comfyanonymous
|
727588d076
|
Fix some new loras.
|
1 year ago |
|
comfyanonymous
|
4f9b6f39d1
|
Fix potential issue with Save Checkpoint.
|
1 year ago |
|
comfyanonymous
|
5f75d784a1
|
Start is now 0.0 and end is now 1.0 for the timestep ranges.
|
1 year ago |
|
comfyanonymous
|
7ff14b62f8
|
ControlNetApplyAdvanced can now define when controlnet gets applied.
|
1 year ago |
|
comfyanonymous
|
d191c4f9ed
|
Add a ControlNetApplyAdvanced node.
The controlnet can be applied to the positive or negative prompt only by
connecting it correctly.
|
1 year ago |
|
comfyanonymous
|
0240946ecf
|
Add a way to set which range of timesteps the cond gets applied to.
|
1 year ago |
|
comfyanonymous
|
22f29d66ca
|
Try to fix memory issue with lora.
|
1 year ago |
|
comfyanonymous
|
67be7eb81d
|
Nodes can now patch the unet function.
|
1 year ago |
|
comfyanonymous
|
12a6e93171
|
Del the right object when applying lora.
|
1 year ago |
|
comfyanonymous
|
78e7958d17
|
Support controlnet in diffusers format.
|
1 year ago |
|
comfyanonymous
|
09386a3697
|
Fix issue with lora in some cases when combined with model merging.
|
1 year ago |
|
comfyanonymous
|
58b2364f58
|
Properly support SDXL diffusers unet with UNETLoader node.
|
1 year ago |
|
comfyanonymous
|
0115018695
|
Print errors and continue when lora weights are not compatible.
|
1 year ago |
|
comfyanonymous
|
0b284f650b
|
Fix typo.
|
1 year ago |
|
comfyanonymous
|
e032ca6138
|
Fix ddim issue with older torch versions.
|
1 year ago |
|
comfyanonymous
|
18885f803a
|
Add MX450 and MX550 to list of cards with broken fp16.
|
1 year ago |
|
comfyanonymous
|
9ba440995a
|
It's actually possible to torch.compile the unet now.
|
1 year ago |
|
comfyanonymous
|
51d5477579
|
Add key to indicate checkpoint is v_prediction when saving.
|
1 year ago |
|
comfyanonymous
|
ff6b047a74
|
Fix device print on old torch version.
|
1 year ago |
|
comfyanonymous
|
9871a15cf9
|
Enable --cuda-malloc by default on torch 2.0 and up.
Add --disable-cuda-malloc to disable it.
|
1 year ago |
|
comfyanonymous
|
55d0fca9fa
|
--windows-standalone-build now enables --cuda-malloc
|
1 year ago |
|
comfyanonymous
|
1679abd86d
|
Add a command line argument to enable backend:cudaMallocAsync
|
1 year ago |
|
comfyanonymous
|
3a150bad15
|
Only calculate randn in some samplers when it's actually being used.
|
1 year ago |
|
comfyanonymous
|
ee8f8ee07f
|
Fix regression with ddim and uni_pc when batch size > 1.
|
1 year ago |
|
comfyanonymous
|
3ded1a3a04
|
Refactor of sampler code to deal more easily with different model types.
|
1 year ago |
|
comfyanonymous
|
5f57362613
|
Lower lora ram usage when in normal vram mode.
|
1 year ago |
|
comfyanonymous
|
490771b7f4
|
Speed up lora loading a bit.
|
1 year ago |
|
comfyanonymous
|
50b1180dde
|
Fix CLIPSetLastLayer not reverting when removed.
|
1 year ago |
|
comfyanonymous
|
6fb084f39d
|
Reduce floating point rounding errors in loras.
|
1 year ago |
|
comfyanonymous
|
91ed2815d5
|
Add a node to merge CLIP models.
|
1 year ago |
|
comfyanonymous
|
b2f03164c7
|
Prevent the clip_g position_ids key from being saved in the checkpoint.
This is to make it match the official checkpoint.
|
1 year ago |
|
comfyanonymous
|
46dc050c9f
|
Fix potential tensors being on different devices issues.
|
1 year ago |
KarryCharon
|
3e2309f149
|
fix mps miss import
|
1 year ago |
|
comfyanonymous
|
606a537090
|
Support SDXL embedding format with 2 CLIP.
|
1 year ago |
|
comfyanonymous
|
6ad0a6d7e2
|
Don't patch weights when multiplier is zero.
|
1 year ago |
|
comfyanonymous
|
d5323d16e0
|
latent2rgb matrix for SDXL.
|
1 year ago |
|
comfyanonymous
|
0ae81c03bb
|
Empty cache after model unloading for normal vram and lower.
|
1 year ago |
|
comfyanonymous
|
d3f5998218
|
Support loading clip_g from diffusers in CLIP Loader nodes.
|
1 year ago |
|
comfyanonymous
|
a9a4ba7574
|
Fix merging not working when model2 of model merge node was a merge.
|
1 year ago |
|
comfyanonymous
|
e7bee85df8
|
Add arguments to run the VAE in fp16 or bf16 for testing.
|
1 year ago |
