comfyanonymous
67892b5ac5
Refactor and improve model_management code related to free memory.
1 year ago
space-nuko
499641ebf1
More accurate total
1 year ago
space-nuko
b5dd15c67a
System stats endpoint
1 year ago
comfyanonymous
5c38958e49
Tweak lowvram model memory so it's closer to what it was before.
1 year ago
comfyanonymous
94680732d3
Empty cache on mps.
1 year ago
comfyanonymous
03da8a3426
This is useless for inference.
1 year ago
comfyanonymous
eb448dd8e1
Auto load model in lowvram if not enough memory.
1 year ago
comfyanonymous
b9818eb910
Add route to get safetensors metadata:
...
/view_metadata/loras?filename=lora.safetensors
1 year ago
comfyanonymous
a532888846
Support VAEs in diffusers format.
2 years ago
comfyanonymous
0fc483dcfd
Refactor diffusers model convert code to be able to reuse it.
2 years ago
comfyanonymous
eb4bd7711a
Remove einops.
2 years ago
comfyanonymous
87ab25fac7
Do operations in same order as the one it replaces.
2 years ago
comfyanonymous
e1278fa925
Support old pytorch versions that don't have weights_only.
2 years ago
BlenderNeko
8b4b0c3188
vecorized bislerp
2 years ago
comfyanonymous
b8ccbec6d8
Various improvements to bislerp.
2 years ago
comfyanonymous
34887b8885
Add experimental bislerp algorithm for latent upscaling.
...
It's like bilinear but with slerp.
2 years ago
comfyanonymous
6cc450579b
Auto transpose images from exif data.
2 years ago
comfyanonymous
dc198650c0
sample_dpmpp_2m_sde no longer crashes when step == 1.
2 years ago
comfyanonymous
069657fbf3
Add DPM-Solver++(2M) SDE and exponential scheduler.
...
exponential scheduler is the one recommended with this sampler.
2 years ago
comfyanonymous
b8636a44aa
Make scaled_dot_product switch to sliced attention on OOM.
2 years ago
comfyanonymous
797c4e8d3b
Simplify and improve some vae attention code.
2 years ago
comfyanonymous
ef815ba1e2
Switch default scheduler to normal.
2 years ago
comfyanonymous
3a1f47764d
Print the torch device that is used on startup.
2 years ago
BlenderNeko
1201d2eae5
Make nodes map over input lists ( #579 )
...
* 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
2 years ago
BlenderNeko
19c014f429
comment out annoying print statement
2 years ago
BlenderNeko
d9e088ddfd
minor changes for tiled sampler
2 years ago
comfyanonymous
f7c0f75d1f
Auto batching improvements.
...
Try batching when cond sizes don't match with smart padding.
2 years ago
comfyanonymous
314e526c5c
Not needed anymore because sampling works with any latent size.
2 years ago
comfyanonymous
c6e34963e4
Make t2i adapter work with any latent resolution.
2 years ago
comfyanonymous
6fc4917634
Make maximum_batch_area take into account python2.0 attention function.
...
More conservative xformers maximum_batch_area.
2 years ago
comfyanonymous
678f933d38
maximum_batch_area for xformers.
...
Remove useless code.
2 years ago
EllangoK
8e03c789a2
auto-launch cli arg
2 years ago
comfyanonymous
cb1551b819
Lowvram mode for gligen and fix some lowvram issues.
2 years ago
comfyanonymous
af9cc1fb6a
Search recursively in subfolders for embeddings.
2 years ago
comfyanonymous
6ee11d7bc0
Fix import.
2 years ago
comfyanonymous
bae4fb4a9d
Fix imports.
2 years ago
comfyanonymous
fcf513e0b6
Refactor.
2 years ago
pythongosssss
5eeecf3fd5
remove unused import
2 years ago
pythongosssss
8912623ea9
use comfy progress bar
2 years ago
comfyanonymous
908dc1d5a8
Add a total_steps value to sampler callback.
2 years ago
pythongosssss
27df74101e
reduce duplication
2 years ago
comfyanonymous
93c64afaa9
Use sampler callback instead of tqdm hook for progress bar.
2 years ago
pythongosssss
06ad35b493
added progress to encode + upscale
2 years ago
comfyanonymous
ba8a4c3667
Change latent resolution step to 8.
2 years ago
comfyanonymous
66c8aa5c3e
Make unet work with any input shape.
2 years ago
comfyanonymous
9c335a553f
LoKR support.
2 years ago
comfyanonymous
d3293c8339
Properly disable all progress bars when disable_pbar=True
2 years ago
BlenderNeko
a2e18b1504
allow disabling of progress bar when sampling
2 years ago
comfyanonymous
071011aebe
Mask strength should be separate from area strength.
2 years ago
Jacob Segal
af02393c2a
Default to sampling entire image
...
By default, when applying a mask to a condition, the entire image will
still be used for sampling. The new "set_area_to_bounds" option on the
node will allow the user to automatically limit conditioning to the
bounds of the mask.
I've also removed the dependency on torchvision for calculating bounding
boxes. I've taken the opportunity to fix some frustrating details in the
other version:
1. An all-0 mask will no longer cause an error
2. Indices are returned as integers instead of floats so they can be
used to index into tensors.
2 years ago