6761233e9d
Implement Self-Attention Guidance ( #2201 )
...
* First SAG test
* need to put extra options on the model instead of patcher
* no errors and results seem not-broken
* Use @ashen-uncensored formula, which works better!!!
* Fix a crash when using weird resolutions. Remove an unnecessary UNet call
* Improve comments, optimize memory in blur routine
* SAG works with sampler_cfg_function
1 year ago
9ac0b487ac
Make --gpu-only put intermediate values in GPU memory instead of cpu.
1 year ago
d19de2753e
Merge branch 'fix_folders_handling' of https://github.com/fazo96/ComfyUI
1 year ago
798a34d009
Lower compress level for image preview.
1 year ago
5b37270d3a
Add a lora loader node for models with no CLIP.
1 year ago
42dfae6331
Nodes to properly use the SDV img2vid checkpoint.
...
The img2vid model is conditioned on clip vision output only which means
there's no CLIP model which is why I added a ImageOnlyCheckpointLoader to
load it. Note that the unClipCheckpointLoader can also load it because it
also has a CLIP_VISION output.
SDV_img2vid_Conditioning is the node used to pass the right conditioning
to the img2vid model.
VideoLinearCFGGuidance applies a linearly decreasing CFG scale to each
video frame from the cfg set in the sampler node to min_cfg.
SDV_img2vid_Conditioning can be found in conditioning->video_models
ImageOnlyCheckpointLoader can be found in loaders->video_models
VideoLinearCFGGuidance can be found in sampling->video_models
1 year ago
1964bf1e78
fix: folder handling issues
1 year ago
cd4fc77d5f
Add taesd and taesdxl to VAELoader node.
...
They will show up if both the taesd_encoder and taesd_decoder or taesdxl
model files are present in the models/vae_approx directory.
1 year ago
8a451234b3
Add ImageCrop node.
1 year ago
bd07ad1861
Add PatchModelAddDownscale (Kohya Deep Shrink) node.
...
By adding a downscale to the unet in the first timesteps this node lets
you generate images at higher resolutions with less consistency issues.
1 year ago
dcec1047e6
Invert the start and end percentages in the code.
...
This doesn't affect how percentages behave in the frontend but breaks
things if you relied on them in the backend.
percent_to_sigma goes from 0 to 1.0 instead of 1.0 to 0 for less confusion.
Make percent 0 return an extremely large sigma and percent 1.0 return a
zero one to fix imprecision.
1 year ago
ca71e542d2
Lower cfg step to 0.1 in sampler nodes.
1 year ago
844dbf97a7
Add: advanced->model->ModelSamplingDiscrete node.
...
This allows changing the sampling parameters of the model (eps or vpred)
or set the model to use zsnr.
1 year ago
1443caf373
HyperTile node, can be found in: _for_testing->HyperTile
1 year ago
d44a2de49f
Make VAE code closer to sgm.
1 year ago
7e09e889e3
Make clear that the old CheckpointLoader is deprecated.
1 year ago
20d3852aa1
Pull some small changes from the other repo.
1 year ago
9212bea87c
Change a few things in #1578 .
1 year ago
ece69bf28c
Change channel type to MASK (reduced redundancy, increased usability)
1 year ago
d06cd2805d
Added support for Porter-Duff image compositing
1 year ago
ec454c771b
Refactor with code from comment of #1588
1 year ago
1adcc4c3a2
Add a SamplerCustom Node.
...
This node takes a list of sigmas and a sampler object as input.
This lets people easily implement custom schedulers and samplers as nodes.
More nodes will be added to it in the future.
1 year ago
bf3fc2f1b7
Refactor sampling related code.
1 year ago
9546a798fb
Make LoadImage and LoadImageMask return masks in batch format.
1 year ago
ba7dfd60f2
Merge branch 'proportional-scale' of https://github.com/jn-jairo/ComfyUI
1 year ago
2381d36e6d
1024 wasn't enough.
1 year ago
42f6d1ebe2
Increase maximum batch sizes of empty image nodes.
1 year ago
77c124c5a1
Fix typo.
1 year ago
593b7069e7
Proportional scale latent and image
1 year ago
fd93c759e2
Implement FreeU: Free Lunch in Diffusion U-Net node.
...
_for_testing->FreeU
1 year ago
422d16c027
Add some nodes to add, subtract and multiply latents.
1 year ago
1122df1a20
Increase range of lora strengths.
1 year ago
2b6b178173
Added support for lanczos scaling
1 year ago
0109431626
Lower the minimum resolution of EmptyLatentImage.
1 year ago
db63aa7e53
Nodes can now control the rounding in the UI.
1 year ago
f88f7f413a
Add a ConditioningSetAreaPercentage node.
1 year ago
36ea8784a8
Only return tuple of 3 args in CheckpointLoaderSimple.
2 years ago
57beace324
Fix VAEDecodeTiled minimum.
2 years ago
7e941f9f24
Clean up DiffusersLoader node.
2 years ago
8ddd081b09
Use the same units for tile size in VAEDecodeTiled and VAEEncodeTiled.
2 years ago
d86b222fe9
Reduce min tile size for encode
2 years ago
9196588088
Make tile size in Tiled VAE encode/decode user configurable
2 years ago
c77f02e1c6
Move controlnet code to comfy/controlnet.py
2 years ago
d7b3b0f8c1
Don't hardcode node names for image upload widget.
2 years ago
9b1d5a587c
Allow loading js extensions without copying to /web folder
2 years ago
27b87c25a1
Add an EmptyImage node.
...
TODO: implement color picker in the frontend.
2 years ago
e7d88855f4
Add node to batch images together.
2 years ago
9cc12c833d
CLIPVisionEncode can now encode multiple images.
2 years ago
fa962e86c1
Make LatentBlend more consistent with other nodes.
2 years ago
d712193885
Add LatentBlend node, allowing for blending between two Latent inputs.
2 years ago
3dcad78fe1
SaveLatent reports its outputs so they are visible to API
2 years ago
1141029a4a
Add --disable-metadata argument to disable saving metadata in files.
2 years ago
5f75d784a1
Start is now 0.0 and end is now 1.0 for the timestep ranges.
2 years ago
7ff14b62f8
ControlNetApplyAdvanced can now define when controlnet gets applied.
2 years ago
d191c4f9ed
Add a ControlNetApplyAdvanced node.
...
The controlnet can be applied to the positive or negative prompt only by
connecting it correctly.
2 years ago
0240946ecf
Add a way to set which range of timesteps the cond gets applied to.
2 years ago
a8f3bbc35d
Patch del self.loaded_lora to prevent error with persistent lora_name swapping
2 years ago
eed3042830
Move conditioning concat node to conditioning section.
2 years ago
bdba394290
Add a canny preprocessor node.
2 years ago
3bc8be33e4
Don't let custom nodes overwrite base nodes.
2 years ago
f4b9390623
Add a random string to the temp prefix for PreviewImage.
2 years ago
ccb1b25908
Add a conditioning concat node.
2 years ago
af7a49916b
Support loading unet files in diffusers format.
2 years ago
3a09fac835
ConditioningAverage now also averages the pooled output.
2 years ago
d94ddd8548
Add text encode nodes to control the extra parameters in SDXL.
2 years ago
5a9ddf94eb
LoraLoader node now caches the lora file between executions.
2 years ago
2d880fec3a
Add a node to zero out the cond to advanced/conditioning
...
The stability streamlit example passes a zero cond as the negative input
so using this for the negative input makes outputs match the streamlit.
2 years ago
9b93b920be
Add CheckpointSave node to save checkpoints.
...
The created checkpoints contain workflow metadata that can be loaded by
dragging them on top of the UI or loading them with the "Load" button.
Checkpoints will be saved in fp16 or fp32 depending on the format ComfyUI
is using for inference on your hardware. To force fp32 use: --force-fp32
Anything that patches the model weights like merging or loras will be
saved.
The output directory is currently set to: output/checkpoints but that might
change in the future.
2 years ago
4eab00e14b
Set the seed in the SDE samplers to make them more reproducible.
2 years ago
cef6aa62b2
Add support for TAESD decoder for SDXL.
2 years ago
20f579d91d
Add DualClipLoader to load clip models for SDXL.
...
Update LoadClip to load clip models for SDXL refiner.
2 years ago
05676942b7
Add some more transformer hooks and move tomesd to comfy_extras.
...
Tomesd now uses q instead of x to decide which tokens to merge because
it seems to give better results.
2 years ago
8607c2d42d
Move latent scale factor from VAE to model.
2 years ago
f87ec10a97
Support base SDXL and SDXL refiner models.
...
Large refactor of the model detection and loading code.
2 years ago
bf3f271775
Add some nodes for basic model merging.
2 years ago
8c9c94b5f3
Add bicubic upscale method.
2 years ago
67833c83d8
Add ImageScaleBy node.
2 years ago
c64ca8c0b2
Refactor unCLIP noise augment out of samplers.py
2 years ago
a3a713b6c5
Refactor previews into one command line argument.
...
Clean up a few things.
2 years ago
8b4a6c19c2
Fix
2 years ago
3e17971acb
preview method autodetection
2 years ago
d5a28fadaa
Add latent2rgb preview
2 years ago
70d72c4336
Slightly less vibrant sample
2 years ago
48f7ec750c
Make previews into cli option
2 years ago
f326a0a468
Make new LATENT_PREVIEWER type for declaring KSampler preview methods
2 years ago
b4f434ee66
Preview sampled images with TAESD
2 years ago
0fc483dcfd
Refactor diffusers model convert code to be able to reuse it.
2 years ago
4d1ed829d9
Don't load some model types if weight is zero
2 years ago
c00bb1a0b7
Add a latent upscale by node.
2 years ago
34887b8885
Add experimental bislerp algorithm for latent upscaling.
...
It's like bilinear but with slerp.
2 years ago
6cc450579b
Auto transpose images from exif data.
2 years ago
8bbd9815a9
Support loading fp16 latent files.
2 years ago
62a371e12b
Load workflow from latent file.
2 years ago
faf899ad5a
LoadLatent and SaveLatent should behave like the LoadImage and SaveImage.
2 years ago
a7375103b9
Some small changes to Load/SaveLatent.
2 years ago
e7f2816c6f
feat:Latent Save/Load ( #662 )
...
* wip
* latent dir
* fix
* fix
* now working
* mark todo
* remove server.py changes to separate PRt
---------
Co-authored-by: Lt.Dr.Data <lt.dr.data@gmail.com>
2 years ago
2ec6d1c6e3
Don't import custom nodes when the folder ends with .disabled
2 years ago
c02a554bcf
Make DiffusersLoader work with subfolders.
2 years ago
acff543d66
Remove useless code.
2 years ago
b0505eb7ab
Return right type when none specified in upload route.
...
Switch time.time to time.perf_counter for custom node import times.
2 years ago
Rafie Walker
comfyanonymous
Enrico Fasoli
MoonRide303
Jairo Correa
BVH
pythongosssss
FuamiCake
ComfyUI-Community
space-nuko
Dr.Lt.Data