comfyanonymous
0eaa34ec5b
Fix regular empty latent image not working with SD3 and custom sampler.
2024-06-12 10:32:34 -04:00
comfyanonymous
321e509e0a
Add link to SD3 example page to README.
2024-06-12 09:48:27 -04:00
comfyanonymous
c8b5e08dc3
Default shift value on SD3 is 3.0
2024-06-12 02:24:39 -04:00
comfyanonymous
1ddf512fdc
Don't auto convert clip and vae weights to fp16 when saving checkpoint.
2024-06-12 01:07:58 -04:00
comfyanonymous
32be358213
Save SD3 modelspec.architecture in CheckpointSave node.
2024-06-12 01:02:07 -04:00
comfyanonymous
694e0b48e0
SD3 better memory usage estimation.
2024-06-12 00:49:00 -04:00
comfyanonymous
69c8d6d8a6
Single and dual clip loader nodes support SD3.
...
You can use the CLIPLoader to use the t5xxl only or the DualCLIPLoader to
use CLIP-L and CLIP-G only for sd3.
2024-06-11 23:27:39 -04:00
comfyanonymous
0e49211a11
Load the SD3 T5xxl model in the same dtype stored in the checkpoint.
2024-06-11 17:03:26 -04:00
comfyanonymous
5889b7ca0a
Support multiple text encoder configurations on SD3.
2024-06-11 13:14:43 -04:00
comfyanonymous
1c34d338d7
Update EmptySD3LatentImage to use 1024 resolution by default.
2024-06-11 07:37:22 -04:00
comfyanonymous
9424522ead
Reuse code.
2024-06-11 07:20:26 -04:00
comfyanonymous
4134564dc1
Require safetensors library to be at least 0.4.2 for fp8 support.
2024-06-11 06:26:13 -04:00
comfyanonymous
a82fae2375
Fix bug with cosxl edit model.
2024-06-10 16:00:03 -04:00
comfyanonymous
8c4a9befa7
SD3 Support.
2024-06-10 14:06:23 -04:00
comfyanonymous
a5e6a632f9
Support sampling non 2D latents.
2024-06-10 01:31:09 -04:00
comfyanonymous
742d5720d1
Support zeroing out text embeddings with the attention mask.
2024-06-09 16:51:58 -04:00
comfyanonymous
6cd8ffc465
Reshape the empty latent image to the right amount of channels if needed.
2024-06-08 02:35:08 -04:00
comfyanonymous
56333d4850
Use the end token for the text encoder attention mask.
2024-06-07 03:05:23 -04:00
comfyanonymous
0dccb4617d
Remove some unnecessary arguments.
2024-06-06 14:49:45 -04:00
comfyanonymous
104fcea0c8
Add function to get the list of currently loaded models.
2024-06-05 23:25:16 -04:00
comfyanonymous
b1fd26fe9e
pytorch xpu should be flash or mem efficient attention?
2024-06-04 17:44:14 -04:00
comfyanonymous
cb8d0ebccc
Don't load the view coordinates when loading a workflow from the history.
...
I think this makes things slightly less annoying for some users.
2024-06-03 19:48:27 -04:00
comfyanonymous
809cc85a8e
Remove useless code.
2024-06-02 19:23:37 -04:00
comfyanonymous
b249862080
Add an annoying print to a function I want to remove.
2024-06-01 12:47:31 -04:00
comfyanonymous
04b308229e
Small refactor of preview code.
2024-05-31 11:18:37 -04:00
comfyanonymous
bf3e334d46
Disable non_blocking when --deterministic or directml.
2024-05-30 11:07:38 -04:00
comfyanonymous
71ec5b144e
Update commands to install nightly pytorch in readme.
2024-05-29 00:20:02 -04:00
comfyanonymous
91542d4f8b
Import spandrel_extra_arches if present.
...
I will not add this dependency to the default ones because models in the
spandrel_extra_arches package are non commercial and therefore not
compatible with free software licenses like the one ComfyUI uses.
If you don't mind this you can install it manually yourself.
2024-05-28 01:42:11 -04:00
comfyanonymous
0920e0e5fe
Remove some unused imports.
2024-05-27 19:08:27 -04:00
comfyanonymous
16a493a190
Keep compatibility with some custom nodes.
2024-05-26 15:37:41 -04:00
comfyanonymous
9a151b7def
Fix issue and unpin spandrel package.
2024-05-26 13:44:47 -04:00
comfyanonymous
ffc4b7c30e
Fix DORA strength.
...
This is a different version of #3298 with more correct behavior.
2024-05-25 02:50:11 -04:00
comfyanonymous
efa5a711b2
Reduce memory usage when applying DORA: #3557
2024-05-24 23:36:48 -04:00
comfyanonymous
58c9838274
Speed up TAESD preview.
2024-05-24 02:37:57 -04:00
comfyanonymous
b02bcced05
Fix FreeU not working when shape is tensor.
2024-05-23 11:48:04 -04:00
comfyanonymous
6507a9c716
Remove the CTRL-Delete keybind.
...
On some keyboards it's apparently too easy to accidentally do CTRL-Delete
when pressing CTRL-Enter repeatedly.
CTRL-Backspace can still be used to clear the workflow.
2024-05-23 01:29:22 -04:00
comfyanonymous
6c23854f54
Fix OSX latent2rgb previews.
2024-05-22 13:56:28 -04:00
comfyanonymous
8508df2569
Work around black image bug on Mac 14.5 by forcing attention upcasting.
2024-05-21 16:56:33 -04:00
comfyanonymous
83d969e397
Disable xformers when tracing model.
2024-05-21 13:55:49 -04:00
comfyanonymous
1900e5119f
Fix potential issue.
2024-05-20 08:19:54 -04:00
comfyanonymous
276f8fce9f
Print error when node is missing.
2024-05-20 07:04:08 -04:00
comfyanonymous
09e069ae6c
Log the pytorch version.
2024-05-20 06:22:29 -04:00
comfyanonymous
11a2ad5110
Fix controlnet not upcasting on models that have it enabled.
2024-05-19 17:58:03 -04:00
comfyanonymous
4ae1515f14
Slightly faster latent2rgb previews.
2024-05-19 17:42:35 -04:00
comfyanonymous
f37a47110b
Make --preview-method auto default to the fast latent2rgb previews.
2024-05-19 11:45:36 -04:00
comfyanonymous
0bdc2b15c7
Cleanup.
2024-05-18 10:11:44 -04:00
comfyanonymous
98f828fad9
Remove unnecessary code.
2024-05-18 09:36:44 -04:00
comfyanonymous
1c4af5918a
Better error message if the webcam node doesn't work.
2024-05-17 14:02:09 -04:00
comfyanonymous
19300655dd
Don't automatically switch to lowvram mode on GPUs with low memory.
2024-05-17 00:31:32 -04:00
comfyanonymous
46daf0a9a7
Add debug options to force on and off attention upcasting.
2024-05-16 04:09:41 -04:00
comfyanonymous
58f8388020
More proper fix for #3484 .
2024-05-16 00:11:01 -04:00
comfyanonymous
2d41642716
Fix lowvram dora issue.
2024-05-15 02:47:40 -04:00
comfyanonymous
ec6f16adb6
Fix SAG.
2024-05-14 18:02:27 -04:00
comfyanonymous
bb4940d837
Only enable attention upcasting on models that actually need it.
2024-05-14 17:00:50 -04:00
comfyanonymous
b0ab31d06c
Refactor attention upcasting code part 1.
2024-05-14 12:47:31 -04:00
comfyanonymous
2de3b69b30
Support saving some more modelspec types.
2024-05-13 21:54:11 -04:00
comfyanonymous
ece5acb8e8
Fix nightly package workflow.
2024-05-12 16:05:10 -04:00
comfyanonymous
794a357f7a
Update the nightly workflow.
2024-05-12 07:24:12 -04:00
comfyanonymous
fa6dd7e5bb
Fix lowvram issue with saving checkpoints.
...
The previous fix didn't cover the case where the model was loaded in
lowvram mode right before.
2024-05-12 06:13:45 -04:00
comfyanonymous
49c20cdc70
No longer necessary.
2024-05-12 05:34:43 -04:00
comfyanonymous
e1489ad257
Fix issue with lowvram mode breaking model saving.
2024-05-11 21:55:20 -04:00
comfyanonymous
4f63ee99f1
Add a button to reset the view.
2024-05-10 17:30:52 -04:00
comfyanonymous
93e876a3be
Remove warnings that confuse people.
2024-05-09 05:29:42 -04:00
comfyanonymous
cd07340d96
Typo fix.
2024-05-08 18:36:56 -04:00
comfyanonymous
c33412288f
Fix issue with loading some JPG: #3416
2024-05-07 05:41:06 -04:00
comfyanonymous
c61eadf69a
Make the load checkpoint with config function call the regular one.
...
I was going to completely remove this function because it is unmaintainable
but I think this is the best compromise.
The clip skip and v_prediction parts of the configs should still work but
not the fp16 vs fp32.
2024-05-06 20:04:39 -04:00
comfyanonymous
565eb6d176
Add a SplitSigmasDenoise node as an alternative to SplitSigmas.
2024-05-05 05:24:36 -04:00
comfyanonymous
72508a8d19
Only set LOAD_TRUNCATED_IMAGES when if the Image open fails.
...
Document which PIL issues this works around.
2024-05-04 03:51:03 -04:00
comfyanonymous
daa92a8ff4
Fix potential issues with the int rounding fix.
2024-05-03 05:49:21 -04:00
comfyanonymous
89d0e9abeb
Fix int widgets rounding.
2024-05-02 03:34:19 -04:00
comfyanonymous
f81a6fade8
Fix some edge cases with samplers and arrays with a single sigma.
2024-05-01 17:05:30 -04:00
comfyanonymous
94d5a12801
Don't load the model in SDTurboScheduler
2024-05-01 16:57:10 -04:00
comfyanonymous
2aed53c4ac
Workaround xformers bug.
2024-04-30 21:23:40 -04:00
comfyanonymous
bb8b48a260
Update Readme.
2024-04-30 20:11:34 -04:00
comfyanonymous
eecd69b53a
Add a SamplerLCMUpscale node.
...
This sampler is an LCM sampler that upscales the latent during sampling.
It can be used to generate at a higher resolution with an LCM model very
quickly.
To try it use it with a basic 5 step LCM workflow with scale_ratio 1.5 or
2.0
2024-04-29 20:00:47 -04:00
comfyanonymous
059773a6df
Add some nodes to multiply the attention in UNet and Clip models.
2024-04-28 13:03:43 -04:00
comfyanonymous
10fcd09f4a
Add a denoise value to AlignYourStepsScheduler.
2024-04-27 00:48:41 -04:00
comfyanonymous
8cab3be673
Update command for AMD stable pytorch install in README.
2024-04-26 15:44:12 -04:00
comfyanonymous
16eabdf70d
Free more vram for upscale models.
2024-04-25 17:04:19 -04:00
comfyanonymous
8dc19e40d1
Don't init a VAE model when there are no VAE weights.
2024-04-24 09:20:31 -04:00
comfyanonymous
27d5808fc4
Increase max lora strength to 100.0
2024-04-23 13:07:39 -04:00
comfyanonymous
d09b5ef4ef
Free some memory before loading upscale models.
2024-04-22 18:51:15 -04:00
comfyanonymous
4ee9aad6ca
Speed up Sharpen node.
2024-04-21 09:02:06 -04:00
comfyanonymous
644a3ae58d
Implement Align Your Steps as a AlignYourStepsScheduler node.
2024-04-20 04:34:12 -04:00
comfyanonymous
133dc3351b
Faster blur.
2024-04-19 03:52:02 -04:00
comfyanonymous
5d08802f78
Sync some minor changes from the other repo.
2024-04-19 03:43:09 -04:00
comfyanonymous
c59fe9f254
Support VAE without quant_conv.
2024-04-18 21:05:33 -04:00
comfyanonymous
d64e217427
Fix annoying float issue causing the value to be rounded to above the max.
2024-04-17 17:34:02 -04:00
comfyanonymous
abc69cab45
Add a helpful warning for links that don't point anywhere.
2024-04-17 12:28:05 -04:00
comfyanonymous
45ec1cbe96
Implement PerpNeg as a guider.
2024-04-16 02:57:34 -04:00
comfyanonymous
8903dce862
This can be removed since PAG doesn't use the uncond.
2024-04-15 12:14:00 -04:00
comfyanonymous
719fb2c81d
Add basic PAG node.
2024-04-14 23:49:50 -04:00
comfyanonymous
258dbc06c3
Fix some memory related issues.
2024-04-14 12:08:58 -04:00
comfyanonymous
744ac944db
Don't make dynamicPrompts the default on multiline string inputs.
...
This should be less confusing to those who want to use multiline input
without them.
2024-04-13 16:18:00 -04:00
comfyanonymous
58812ab8ca
Support SDXS 512 model.
2024-04-12 22:12:35 -04:00
comfyanonymous
0256e7f769
Fix tests.
2024-04-12 20:02:53 -04:00
comfyanonymous
4bd7d55b90
Add some colors to SamplerCustom links.
...
If you don't like them I am open to a PR.
2024-04-11 22:43:05 -04:00
comfyanonymous
fd7c636680
Add an AddNoise node to add noise depending on the sigma.
2024-04-10 23:40:31 -04:00
comfyanonymous
831511a1ee
Fix issue with sampling_settings persisting across models.
2024-04-09 23:20:43 -04:00
comfyanonymous
4201181b35
Add ModelMergeSD1, ModelMergeSD2 and ModelMergeSDXL.
2024-04-09 04:31:14 -04:00
comfyanonymous