cb7c4b4be3
Workaround for lora OOM on lowvram mode.
2024-08-07 14:30:54 -04:00
1208863eca
Fix "Comfy" lora keys.
...
They are in this format now:
diffusion_model.full.model.key.name.lora_up.weight
2024-08-07 13:49:31 -04:00
e1c528196e
Fix bundled embed.
2024-08-07 13:30:45 -04:00
17030fd4c0
Support for "Comfy" lora format.
...
The keys are just: model.full.model.key.name.lora_up.weight
It is supported by all comfyui supported models.
Now people can just convert loras to this format instead of having to ask
for me to implement them.
2024-08-07 13:18:32 -04:00
c19dcd362f
Controlnet code refactor.
2024-08-07 12:59:28 -04:00
1c08bf35b4
Support format for embeddings bundled in loras.
2024-08-07 03:45:25 -04:00
2a02546e20
Add type hints to folder_paths.py ( #4191 )
...
* add type hints to folder_paths.py
* replace deprecated standard collections type hints
* fix type error when using Python 3.8
2024-08-06 21:59:34 -04:00
b334605a66
Fix OOMs happening in some cases.
...
A cloned model patcher sometimes reported a model was loaded on a device
when it wasn't.
2024-08-06 13:36:04 -04:00
de17a9755e
Unload all models if there's an OOM error.
2024-08-06 03:30:28 -04:00
c14ac98fed
Unload models and load them back in lowvram mode no free vram.
2024-08-06 03:22:39 -04:00
2894511893
Clone taesd with depth of 1 to reduce download size. ( #4232 )
2024-08-06 01:46:09 -04:00
f3bc40223a
Add format metadata to CLIP save to make compatible with diffusers safetensors loading ( #4233 )
2024-08-06 01:45:24 -04:00
841e74ac40
Change browser test CI python to 3.8 ( #4234 )
2024-08-06 01:27:28 -04:00
2d75df45e6
Flux tweak memory usage.
2024-08-05 21:58:28 -04:00
1abc9c8703
Stable release uses cached dependencies ( #4231 )
...
* Release stable based on existing tag.
* Update default cuda to 12.1.
2024-08-05 20:07:16 -04:00
8edbcf5209
Improve performance on some lowend GPUs.
2024-08-05 16:24:04 -04:00
e545a636ba
This probably doesn't work anymore.
2024-08-05 12:31:42 -04:00
33e5203a2a
Don't cache index.html ( #4211 )
2024-08-05 12:25:28 -04:00
a178e25912
Fix Flux FP64 math on XPU ( #4210 )
2024-08-05 01:26:20 -04:00
78e133d041
Support simple diffusers Flux loras.
2024-08-04 22:05:48 -04:00
7afa985fba
Correct spelling 'token_weight_pars_t5' to 'token_weight_pairs_t5' ( #4200 )
2024-08-04 17:10:02 -04:00
ddb6a9f47c
Set the step in EmptySD3LatentImage to 16.
...
These models work better when the res is a multiple of 16.
2024-08-04 15:59:02 -04:00
3b71f84b50
ONNX tracing fixes.
2024-08-04 15:45:43 -04:00
0a6b008117
Fix issue with some custom nodes.
2024-08-04 10:03:33 -04:00
56f3c660bf
ModelSamplingFlux now takes a resolution and adjusts the shift with it.
...
If you want to sample Flux dev exactly how the reference code does use
the same resolution as your image in this node.
2024-08-04 04:06:00 -04:00
f7a5107784
Fix crash.
2024-08-03 16:55:38 -04:00
91be9c2867
Tweak lowvram memory formula.
2024-08-03 16:44:50 -04:00
03c5018c98
Lower lowvram memory to 1/3 of free memory.
2024-08-03 15:14:07 -04:00
2ba5cc8b86
Fix some issues.
2024-08-03 15:06:40 -04:00
1e68002b87
Cap lowvram to half of free memory.
2024-08-03 14:50:20 -04:00
ba9095e5bd
Automatically use fp8 for diffusion model weights if:
...
Checkpoint contains weights in fp8.
There isn't enough memory to load the diffusion model in GPU vram.
2024-08-03 13:45:19 -04:00
f123328b82
Load T5 in fp8 if it's in fp8 in the Flux checkpoint.
2024-08-03 12:39:33 -04:00
63a7e8edba
More aggressive batch splitting.
2024-08-03 11:53:30 -04:00
0eea47d580
Add ModelSamplingFlux to experiment with the shift value.
...
Default shift on Flux Schnell is 0.0
2024-08-03 03:54:38 -04:00
7cd0cdfce6
Add advanced model merge node for Flux model.
2024-08-02 23:20:53 -04:00
ea03c9dcd2
Better per model memory usage estimations.
2024-08-02 18:09:24 -04:00
3a9ee995cf
Tweak regular SD memory formula.
2024-08-02 17:34:30 -04:00
47da42d928
Better Flux vram estimation.
2024-08-02 17:02:35 -04:00
17bbd83176
Fix bug loading flac workflow when it contains = character.
2024-08-02 13:14:28 -04:00
bfb52de866
Lower SAG scale step for finer control ( #4158 )
...
* Lower SAG step for finer control
Since the introduction of cfg++ which uses very low cfg value, a step of 0.1 in SAG might be too high for finer control. Even SAG of 0.1 can be too high when cfg is only 0.6, so I change the step to 0.01.
* Lower PAG step as well.
* Update nodes_sag.py
2024-08-02 10:29:03 -04:00
eca962c6da
Add FluxGuidance node.
...
This lets you adjust the guidance on the dev model which is a parameter
that is passed to the diffusion model.
2024-08-02 10:25:49 -04:00
c1696cd1b5
Add missing import ( #4174 )
2024-08-02 09:34:12 -04:00
369f459b20
Fix no longer working on old pytorch.
2024-08-01 22:20:24 -04:00
ce9ac2fe05
Fix clip_g/clip_l mixup ( #4168 )
2024-08-01 21:40:56 -04:00
e638f2858a
Hack to make all resolutions work on Flux models.
2024-08-01 21:39:18 -04:00
a531001cc7
Add CLIPTextEncodeFlux.
2024-08-01 18:53:25 -04:00
d420bc792a
Tweak the memory usage formulas for Flux and SD.
2024-08-01 17:53:45 -04:00
d965474aaa
Make ComfyUI split batches a higher priority than weight offload.
2024-08-01 16:39:59 -04:00
1c61361fd2
Fast preview support for Flux.
2024-08-01 16:28:11 -04:00
a6decf1e62
Fix bfloat16 potentially not being enabled on mps.
2024-08-01 16:18:44 -04:00
comfyanonymous
PhilWun
Robin Huang
Silver
Chenlei Hu
bymyself
a-One-Fan
fgdfgfthgr-fox
Jairo Correa
Alexander Brown