86721d5158
Enable highvram automatically when vram >> ram
2 years ago
75fa162531
Remove sample_ from some sampler names.
...
Old workflows will still work.
2 years ago
9f4214e534
Preparing to add another function to load checkpoints.
2 years ago
3cd7d84b53
Fix uni_pc sampler not working with 1 or 2 steps.
2 years ago
dfb397e034
Fix multiple controlnets not working.
2 years ago
af3cc1b5fb
Fixed issue when batched image was used as a controlnet input.
2 years ago
d2da346b0b
Fix missing variable.
2 years ago
4e6b83a80a
Add a T2IAdapterLoader node to load T2I-Adapter models.
...
They are loaded as CONTROL_NET objects because they are similar.
2 years ago
fcb25d37db
Prepare for t2i adapter.
2 years ago
cf5a211efc
Remove some useless imports
2 years ago
87b00b37f6
Added an experimental VAEDecodeTiled.
...
This decodes the image with the VAE in tiles which should be faster and
use less vram.
It's in the _for_testing section so I might change/remove it or even
add the functionality to the regular VAEDecode node depending on how
well it performs which means don't depend too much on it.
2 years ago
62df8dd62a
Add a node to load diff controlnets.
2 years ago
f04dc2c2f4
Implement DDIM sampler.
2 years ago
2976c1ad28
Uni_PC: make max denoise behave more like other samplers.
...
On the KSamplers denoise of 1.0 is the same as txt2img but there was a
small difference on UniPC.
2 years ago
c9daec4c89
Remove prints that are useless when xformers is enabled.
2 years ago
a7328e4945
Add uni_pc bh2 variant.
2 years ago
d80af7ca30
ControlNetApply now stacks.
...
It can be used to apply multiple control nets at the same time.
2 years ago
00a9189e30
Support old pytorch.
2 years ago
137ae2606c
Support people putting commas after the embedding name in the prompt.
2 years ago
2326ff1263
Add: --highvram for when you want models to stay on the vram.
2 years ago
09f1d76ed8
Fix an OOM issue.
2 years ago
d66415c021
Low vram mode for controlnets.
2 years ago
220a72d36b
Use fp16 for fp16 control nets.
2 years ago
6135a21ee8
Add a way to control controlnet strength.
2 years ago
4efa67fa12
Add ControlNet support.
2 years ago
bc69fb5245
Use inpaint models the proper way by using VAEEncodeForInpaint.
2 years ago
cef2cc3cb0
Support for inpaint models.
2 years ago
07db00355f
Add masks to samplers code for inpainting.
2 years ago
e3451cea4f
uni_pc now works with KSamplerAdvanced return_with_leftover_noise.
2 years ago
f542f248f1
Show the right amount of steps in the progress bar for uni_pc.
...
The extra step doesn't actually call the unet so it doesn't belong in
the progress bar.
2 years ago
f10b8948c3
768-v support for uni_pc sampler.
2 years ago
ce0aeb109e
Remove print.
2 years ago
5489d5af04
Add uni_pc sampler to KSampler* nodes.
2 years ago
1a4edd19cd
Fix overflow issue with inplace softmax.
2 years ago
509c7dfc6d
Use real softmax in split op to fix issue with some images.
2 years ago
7e1e193f39
Automatically enable lowvram mode if vram is less than 4GB.
...
Use: --normalvram to disable it.
2 years ago
324273fff2
Fix embedding not working when on new line.
2 years ago
1f6a467e92
Update ldm dir with latest upstream stable diffusion changes.
2 years ago
773cdabfce
Same thing but for the other places where it's used.
2 years ago
df40d4f3bf
torch.cuda.OutOfMemoryError is not present on older pytorch versions.
2 years ago
e8c499ddd4
Split optimization for VAE attention block.
2 years ago
5b4e312749
Use inplace operations for less OOM issues.
2 years ago
3fd87cbd21
Slightly smarter batching behaviour.
...
Try to keep batch sizes more consistent which seems to improve things on
AMD GPUs.
2 years ago
bbdcf0b737
Use relative imports for k_diffusion.
2 years ago
708138c77d
Remove print.
2 years ago
047775615b
Lower the chances of an OOM.
2 years ago
853e96ada3
Increase it/s by batching together some stuff sent to unet.
2 years ago
c92633eaa2
Auto calculate amount of memory to use for --lowvram
2 years ago
534736b924
Add some low vram modes: --lowvram and --novram
2 years ago
a84cd0d1ad
Don't unload/reload model from CPU uselessly.
2 years ago
b1a7c9ebf6
Embeddings/textual inversion support for SD2.x
2 years ago
1de5aa6a59
Add a CLIPLoader node to load standalone clip weights.
...
Put them in models/clip
2 years ago
56d802e1f3
Use transformers CLIP instead of open_clip for SD2.x
...
This should make things a bit cleaner.
2 years ago
bf9ccffb17
Small fix for SD2.x loras.
2 years ago
678105fade
SD2.x CLIP support for Loras.
2 years ago
ef90e9c376
Add a LoraLoader node to apply loras to models and clip.
...
The models are modified in place before being used and unpatched after.
I think this is better than monkeypatching since it might make it easier
to use faster non pytorch unet inference in the future.
2 years ago
69df7eba94
Add KSamplerAdvanced node.
...
This node exposes more sampling options and makes it possible for example
to sample the first few steps on the latent image, do some operations on it
and then do the rest of the sampling steps. This can be achieved using the
start_at_step and end_at_step options.
2 years ago
1daccf3678
Run softmax in place if it OOMs.
2 years ago
f73e57d881
Add support for textual inversion embedding for SD1.x CLIP.
2 years ago
50db297cf6
Try to fix OOM issues with cards that have less vram than mine.
2 years ago
73f60740c8
Slightly cleaner code.
2 years ago
0108616b77
Fix issue with some models.
2 years ago
2973ff24c5
Round CLIP position ids to fix float issues in some checkpoints.
2 years ago
c4b02059d0
Add ConditioningSetArea node.
...
to apply conditioning/prompts only to a specific area of the image.
Add ConditioningCombine node.
so that multiple conditioning/prompts can be applied to the image at the
same time
2 years ago
acdc6f42e0
Fix loading some malformed checkpoints?
2 years ago
051f472e8f
Fix sub quadratic attention for SD2 and make it the default optimization.
2 years ago
220afe3310
Initial commit.
2 years ago
comfyanonymous