7dde12e5cf
Correct commit_date type to 'int'
2024-08-24 21:17:08 -04:00
86e3ccaa9d
checkpoint change ( #1427 )
...
* checkpoint change
credit to @dermesut
https://github.com/lllyasviel/stable-diffusion-webui-forge/issues/1421
Enables `sd_model_checkpoint` parameter for API payloads
* Omit redundant .save()
2024-08-23 09:01:30 -07:00
df598c4df5
add "Diffusion in Low Bits" to info text
2024-08-22 23:23:14 -07:00
0bf68c5448
revise NGMS
2024-08-22 23:07:05 -07:00
7b3b5fd0fd
ngms ( #1434 )
2024-08-22 23:06:17 -07:00
d169cd5881
Gets type ( #1410 )
2024-08-22 12:50:09 -07:00
08f7487590
revise "send distilled_cfg_scale when generating hires conds"
2024-08-22 06:35:13 -07:00
bfd7fb1d9f
send distilled_cfg_scale when generating hires conds ( #1403 )
2024-08-22 06:33:48 -07:00
852e88564c
Update api.py ( #1389 )
...
get_sd_models now also returning config - fixes the following error:
fastapi.exceptions.ResponseValidationError: 1 validation errors:
{'type': 'missing', 'loc': ('response', 0, 'config'), 'msg': 'Field required', 'input': {'title': 'realisticVisionV51_v51VAE.safetensors', 'model_name': 'realisticVisionV51_v51VAE', 'hash': None, 'sha256': None, 'filename': 'C:\\Users\\Igor\\Documents\\StableProjectorz\\stable-diffusion-webui-forge\\webui\\models\\Stable-diffusion\\realisticVisionV51_v51VAE.safetensors'}, 'url': 'https://errors.pydantic.dev/2.8/v/missing '}
2024-08-21 22:47:40 -07:00
8b90bd88b5
fix #1367
2024-08-20 22:35:25 -07:00
a46cfa6a1d
minor fixes related to Extras tab ( #1312 )
...
* update ui.js - correct index for extras tab
one character change
* Update postprocessing.py
fix missing attribute orig_name by using name instead
avoid duplication of postprocessing text. Previously written twice, to png sections postprocessing and extras.
* Update postprocessing.py
unnecessary line
2024-08-20 20:28:42 -07:00
b2353a4911
fix hires-fix button ( #1360 )
...
underlying gallery object changed with gradio update, old code broken, new code more simple
added check for attempt to upscale grid
removed redundant check already covered by second assert
2024-08-20 20:28:07 -07:00
c127e60cf0
Update sd_samplers_kdiffusion.py
...
add new samplers here
2024-08-20 15:01:58 +01:00
14ac95f908
fix
2024-08-20 01:37:01 -07:00
9f5a27ca4e
Add Align Your Steps GITS, AYS 11 Steps and AYS 32 Steps Schedulers.
2024-08-19 16:57:58 -04:00
0b70b7287c
gradio
2024-08-19 09:12:38 -07:00
584b6c998e
#1294
2024-08-19 09:09:22 -07:00
96f264ec6a
add a way to save models
2024-08-19 06:30:49 -07:00
d38e560e42
Implement some rethinking about LoRA system
...
1. Add an option to allow users to use UNet in fp8/gguf but lora in fp16.
2. All FP16 loras do not need patch. Others will only patch again when lora weight change.
3. FP8 unet + fp16 lora are available (somewhat only available) in Forge now. This also solves some “LoRA too subtle” problems.
4. Significantly speed up all gguf models (in Async mode) by using independent thread (CUDA stream) to compute and dequant at the same time, even when low-bit weights are already on GPU.
5. View “online lora” as a module similar to ControlLoRA so that it is moved to GPU together with model when sampling, achieving significant speedup and perfect low VRAM management simultaneously.
2024-08-19 04:31:59 -07:00
22a19943d2
fix lcm
2024-08-18 19:23:12 -07:00
93bfd7f85b
invalidate cond cache if distilled CFG changed ( #1240 )
...
* Update processing.py
add distilled_cfg_scale to params that invalidate cond cache
* Update ui.py
distilled CFG and CFG step size 0.1 (from 0.5)
2024-08-17 19:34:11 -07:00
0f266c48bd
make clear_prompt_cache a function
2024-08-17 19:00:13 -07:00
93b40f355e
Forge Space and BiRefNet
2024-08-17 08:29:08 -07:00
1bd6cf0e0c
Support LoRAs for Q8/Q5/Q4 GGUF Models
...
what a crazy night of math
2024-08-15 05:34:46 -07:00
447f261154
fix
2024-08-15 01:56:21 -07:00
3acb50c40e
integrate llama3's GGUF
2024-08-15 01:45:29 -07:00
d8b83a9501
gguf preview
2024-08-15 00:03:32 -07:00
dad6e9ad02
Fix type error in BLIP model generation by converting configuration values to integers. ( #1113 )
...
- Cast `num_beams` and `min_length` to integers in `interrogate.py` when passing them to the BLIP model's `generate` function.
- Resolves issues with type mismatches that caused crashes during the interrogation process.
2024-08-14 17:37:12 -07:00
4eadf1b5e6
Fix validation error by updating SDModelItem config field to be optional. Changed config field in SDModelItem model to Optional[str] with default value None. ( #1096 )
2024-08-14 17:35:17 -07:00
b09c24ef51
add fp16_fix
2024-08-14 17:10:03 -07:00
bb58520a4c
completely solve 'NoneType' object is not iterable
2024-08-13 15:36:18 -07:00
61f83dd610
support all flux models
2024-08-13 05:42:17 -07:00
d08c9c809c
cleanup codes
2024-08-11 22:07:54 -07:00
cd8a181438
make flux inpaint and soft inpaint works
2024-08-11 18:18:16 -07:00
36250c78f5
fix flux img2img and highres.fix math
2024-08-11 18:07:50 -07:00
19b41b9438
Add option to experiment with results from other impl
...
Setting -> Compatibility -> Try to reproduce the results from external software
2024-08-11 17:02:50 -07:00
08958f84c0
do not raise
2024-08-11 16:21:52 -07:00
fec56de319
show xformers on UI
2024-08-11 15:58:30 -07:00
8af421d53a
remove old codes
2024-08-11 15:55:07 -07:00
261b7e7b8b
update version check
...
but they are not used - weird a1111
2024-08-11 15:48:47 -07:00
9001968898
Change default combination to a relatively safe one
2024-08-11 14:26:34 -07:00
86ee2d9462
improve hints
2024-08-11 02:00:53 -07:00
cfa5242a75
forge 2.0.0
...
see also discussions
2024-08-10 19:24:19 -07:00
593455c4de
global unload after env var change
2024-08-08 22:34:55 -07:00
d3b81924df
make none default to clip
2024-08-08 18:12:06 -07:00
cebd828df2
backward compatibility
2024-08-08 17:39:01 -07:00
20e1ba4a82
fix
2024-08-08 15:08:20 -07:00
6921420b3f
Load Model only when click Generate
...
#964
2024-08-08 14:51:13 -07:00
ce3f0f86b4
prompt-all-in-one now works
...
fix for webui backward compatibility
2024-08-08 13:46:42 -07:00
1802850eb1
disable negative prompt UI when CFG==1
2024-08-08 02:24:04 -07:00
altoiddealer