 AUTOMATIC1111
|
d686e73daa
support for SD3: infinite prompt length, token counting
|
il y a 1 an |
 Silver
|
91ecc750be
Update sd_hijack_clip.py
|
il y a 1 an |
 Silver
|
00e09382cd
Add option to enable clip skip for clip L on SDXL
|
il y a 1 an |
|
AUTOMATIC1111
|
eee46a5094
Merge pull request #14981 from wangshuai09/gpu_info_for_ascend
|
il y a 1 an |
 Aarni Koskela
|
e3fa46f26f
Fix various typos with crate-ci/typos
|
il y a 1 an |
 wangshuai09
|
ba66cf8d69
update
|
il y a 1 an |
|
AUTOMATIC1111
|
e2b19900ec
add infotext entry for emphasis; put emphasis into a separate file, add an option to parse but still ignore emphasis
|
il y a 1 an |
 hako-mikan
|
c3c88ca8b4
Update sd_hijack_clip.py
|
il y a 1 an |
|
hako-mikan
|
6b3f7039b6
add option
|
il y a 1 an |
 w-e-w
|
f56a309432
fix missing TI hash
|
il y a 2 ans |
|
AUTOMATIC1111
|
6f0abbb71a
textual inversion support for SDXL
|
il y a 2 ans |
|
AUTOMATIC1111
|
89e6dfff71
repair SDXL
|
il y a 2 ans |
|
AUTOMATIC1111
|
8284ebd94c
fix autograd which i broke for no good reason when implementing SDXL
|
il y a 2 ans |
|
AUTOMATIC1111
|
0198eaec45
Merge pull request #11757 from AUTOMATIC1111/sdxl
|
il y a 2 ans |
|
AUTOMATIC1111
|
2b1bae0d75
add textual inversion hashes to infotext
|
il y a 2 ans |
|
AUTOMATIC1111
|
594c8e7b26
fix CLIP doing the unneeded normalization
|
il y a 2 ans |
|
AUTOMATIC1111
|
da464a3fb3
SDXL support
|
il y a 2 ans |
|
Aarni Koskela
|
51864790fd
Simplify a bunch of `len(x) > 0`/`len(x) == 0` style expressions
|
il y a 2 ans |
|
AUTOMATIC
|
3ec7b705c7
suggestions and fixes from the PR
|
il y a 2 ans |
|
AUTOMATIC
|
a5121e7a06
fixes for B007
|
il y a 2 ans |
|
AUTOMATIC
|
8e2aeee4a1
add BREAK keyword to end current text chunk and start the next
|
il y a 2 ans |
 brkirch
|
df3b31eb55
In-place operations can break gradient calculation
|
il y a 2 ans |
|
AUTOMATIC
|
1740c33547
more comments
|
il y a 2 ans |
|
AUTOMATIC
|
08066676a4
make it not break on empty inputs; thank you tarded, we are
|
il y a 2 ans |
|
AUTOMATIC
|
79e39fae61
CLIP hijack rework
|
il y a 2 ans |
|
AUTOMATIC
|
210449b374
fix 'RuntimeError: Expected all tensors to be on the same device' error preventing models from loading on lowvram/medvram.
|
il y a 2 ans |
|
AUTOMATIC
|
f34c734172
alt-diffusion integration
|
il y a 2 ans |
 zhaohu xing
|
52cc83d36b
fix bugs
|
il y a 2 ans |
|
AUTOMATIC
|
ce6911158b
Add support Stable Diffusion 2.0
|
il y a 2 ans |
