chengzeyi stable-fast issues

chengzeyi / stable-fast

Best inference performance optimization framework for HuggingFace Diffusers on NVIDIA GPUs.

MIT License

1.19k stars 73 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

compile

#166 zeng121 opened 3 weeks ago
0
compile error

#165 zeng121 opened 3 weeks ago
0
compile error

#164 zeng121 opened 3 weeks ago
0
Is stable fast support aingdiffusion_v12 ？

#163 youthbupt opened 1 month ago
0
support hidiffusion?

#162 fhlt opened 2 months ago
0
cuda11.8 how to install stable-fast

#161 mmlingyu opened 2 months ago
1
[Feature Request] Flux model support

#160 Foul-Tarnished opened 3 months ago
4
segfault when latent height or width is not divisible by 4

#159 Birch-san opened 3 months ago
0
Swapping between two different AttentionProcessors

#158 Birch-san opened 3 months ago
1
Doubts about geglu implementation

#157 capyun opened 3 months ago
1
Error occurs when enabling xformers

#156 wxiwnd closed 3 months ago
1
How to debug witch sfast when encountering "segmentation fault"

#155 passerby233 closed 3 months ago
1
支持 sd3

#154 CallmeZhangChenchen opened 4 months ago
9
`RuntimeError: std::bad_alloc` Error occur in other pixel ratio (none-square generation)

#153 Suprhimp opened 4 months ago
4
Switching ControlNet and LoRAs

#152 haihp02 opened 5 months ago
0
Stable-Diffusion-3 Comming

#151 CallmeZhangChenchen closed 4 months ago
0
Is stable-fast is available for use in SD WebUi?

#150 bigmover opened 5 months ago
0
when run powerpaint with brushnet it can't suceess ant print logs

#149 liuchenbaidu opened 6 months ago
1
Would stable-fast support (really) old models?

#148 TheodoreGalanos opened 6 months ago
0
gcc "-std=99" may need to be declared in code

#147 Fengshawn opened 6 months ago
0
Stable-fast compatibility with lightning models

#146 Davin05 opened 6 months ago
1
Install fatal error C1083: windows11

#145 Nucleon729 closed 6 months ago
7
请问一下谁可以共享一下在config.enable_cuda_graph = True的情况下切换Lora已经可用的Demo嘛

#144 CallmeZhangChenchen closed 4 months ago
7
enable_cuda_graph 影响Lora的切换

#143 CallmeZhangChenchen closed 7 months ago
1
Support IP-Adapters

#142 Muawizodux opened 7 months ago
0
opening enable_cuda_graph only support two different size image when reference

#141 zhangp365 opened 7 months ago
2
[Bug] MemoryEfficientAttentionTritonFwdFlashBwOp no longer availale

#140 imD-5 opened 7 months ago
1
Changing working device will cause problem

#139 xiyanjoy closed 7 months ago
3
切换ControlNets

#137 callmezhangchenchenokay closed 7 months ago
1
Fix xformers to work on v0.0.22 - 0.0.25

#136 jonluca closed 8 months ago
2
xformers 25 removes Triton fmha and completely breaks stable-fast, with fix (MemoryEfficientAttentionTritonFwdFlashBwOp, TritonFlashAttentionOp)

#135 tau0-deltav closed 8 months ago
0
Triton group normalization is slower than torch ops in some case.

#134 chenly15 opened 8 months ago
1
v1.0.4 release contains the prebuilt binary for 1.0.3

#133 Zxilly closed 8 months ago
0
can stable fast used with diffusers device_map='auto'

#132 zhangvia opened 8 months ago
1
Issue of too much recompilation after modification of cross attention processor

#131 xziayro opened 8 months ago
1
How can I load the whole model from the compiled one instead of loading only the unet to current existing sd model?

#130 quocanh34 opened 8 months ago
0
Installation failed with an error

#129 cavaniho opened 9 months ago
2
FP8 support in stable fast

#128 jkrauss82 opened 9 months ago
6
accelent the stablediffusion with lora and controlnet?

#127 henbucuoshanghai opened 9 months ago
2
Inference Error for with Dynamix Resolution

#126 alecyan1993 opened 9 months ago
1
Possible support for stable-cascade

#125 JohnnyRacer opened 9 months ago
4
[Feature Request] Support for webui-forge?

#124 banjaminicc opened 9 months ago
1
any gains to be had on sdxl-turbo?

#123 vibber opened 9 months ago
1
Tensor Dimension Mismatch

#122 alecyan1993 opened 9 months ago
1
Inpainting only works with 512x512 using stable-fast

#121 lunatolun opened 9 months ago
4
Can we use some of the optimization functions on other models unrelated to SD and diffusion ?

#120 oxysoft opened 9 months ago
1
Getting This To Work With SD.Next

#119 gohan2091 closed 9 months ago
18
compile vae causes segmentation fault at deconstruction.

#118 turbo0628 closed 9 months ago
0
Cuda OOM issues

#117 alecyan1993 opened 9 months ago
4
README: Add note about xformers with PyTorch < 2.2.0

#116 yondonfu opened 9 months ago
0