issues
search
ogkalu2
/
Merge-Stable-Diffusion-models-without-distortion
Adaptation of the merging method described in the paper - Git Re-Basin: Merging Models modulo Permutation Symmetries (https://arxiv.org/abs/2209.04836) for Stable Diffusion
MIT License
140
stars
21
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Add SD2.1 support (and remove rant)
#48
6DammK9
closed
2 months ago
1
Feature request (Dreaming): Algorithm 3 "MERGEMANY"
#47
6DammK9
opened
2 months ago
0
Real SDXL support, added ETA, and some ranting
#46
6DammK9
closed
2 months ago
0
Weird bg connections?
#45
Xynonners
opened
3 months ago
2
Writing a PermutationSpec for SDXL
#44
AI-Casanova
opened
8 months ago
8
set special_layer
#43
wkpark
closed
8 months ago
0
fix fp32 bug
#42
wkpark
closed
9 months ago
0
typo
#41
wkpark
closed
9 months ago
0
add --fixclip optional
#40
wkpark
closed
10 months ago
0
Misc fixes
#39
wkpark
closed
11 months ago
0
SD 2.x model merge broken (or just not implemented yet)
#38
gxmikvid
opened
12 months ago
3
license?
#37
s1dlx
closed
1 year ago
2
KeyError
#36
robot-never-die
opened
1 year ago
4
Please Help,can't use it!
#35
xsmilingtoast
closed
1 year ago
0
Please help i can;t use it!
#34
xsmilingtoast
closed
1 year ago
0
Please help i can't use it
#33
xsmilingtoast
closed
1 year ago
0
torch.load can be unsafe due to pickle exploits
#32
zwishenzug
closed
11 months ago
1
Code expects models to have 'state_dict'
#31
zwishenzug
closed
11 months ago
1
dot not implemented for half error
#30
zwishenzug
closed
11 months ago
0
Merging breaks position_ids in CLIP model
#29
zwishenzug
closed
11 months ago
0
KeyError: 'cond_stage_model.transformer.text_model.embeddings.position_ids'
#28
R-N
opened
1 year ago
5
float32 mode broken?
#27
wilhelmw201
closed
9 months ago
2
Can You turn it into extension for auto1111 webui ?
#26
2blackbar
opened
1 year ago
0
How many iteraions are needed to merge two models?
#25
catchstar
opened
1 year ago
1
Are models being generated with the correct alpha?
#24
brucethemoose
opened
1 year ago
1
Request for Optimization, fixing issues with CUDA devices
#23
LumiWasTaken
opened
1 year ago
13
Torch 12+
#22
brucethemoose
closed
1 year ago
2
Request: Can the code be modified to merge SD 1.x and SD 2.1 checkpoints?
#21
gessyoo
opened
1 year ago
1
RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu! (when checking argument for argument index in method wrapper__index_select)
#20
T0b1maru
closed
1 year ago
2
KeyError: 'model_ema.diffusion_modeltime_embed0weight'
#19
NaughtDZ
opened
1 year ago
3
add support for safentensors
#18
User1231300
closed
11 months ago
1
Requirements and windows one-click venv install
#17
ClashSAN
closed
1 year ago
0
Slightly faster. Changed weighted permutations to be less extreme usi…
#16
vaguenebula
closed
1 year ago
1
KeyError: 'embedding_manager.embedder.transformer.text_model.embeddings.position_ids'
#15
D-Ogi
opened
1 year ago
6
Can't use in colab?
#14
User1231300
opened
1 year ago
5
Division by 0 error
#13
thojmr
closed
1 year ago
1
Google Colab version pointing to Google Drive folder for ckpts?
#12
bennitheblog
opened
1 year ago
5
Fix divide by zero
#11
vaguenebula
closed
1 year ago
0
Fix for model bias
#10
vaguenebula
closed
1 year ago
0
Gradual sum of models. Added fp16 and iterations options.
#9
vaguenebula
closed
1 year ago
0
Bias toward model b
#8
vaguenebula
closed
1 year ago
7
Changed to half precision (temporary). Use GPU for matrix multiplicat…
#7
vaguenebula
closed
1 year ago
28
KeyError: 'log_one_minus_alphas_cumprod'
#6
7Tenku
opened
1 year ago
5
RuntimeError: expected scalar type Float but found Half
#5
thojmr
closed
1 year ago
3
device for weight_matching permutation
#4
Pyr-000
opened
1 year ago
0
specifying a custom device crashes the merge
#3
Pyr-000
opened
1 year ago
1
Update SD_rebasin_merge.py
#2
Dawgmastah
opened
1 year ago
0
After running for a while, get error
#1
Dawgmastah
opened
1 year ago
279