chengzeyi/stable-fast
Best inference performance optimization framework for HuggingFace Diffusers on NVIDIA GPUs.
PythonMIT
Issues
- 3
[Feature Request] Flux model support
#160 opened by Foul-Tarnished - 0
- 0
compile error
#165 opened by zeng121 - 0
compile error
#164 opened by zeng121 - 0
Is stable fast support aingdiffusion_v12 ?
#163 opened by youthbupt - 0
support hidiffusion?
#162 opened by fhlt - 1
cuda11.8 how to install stable-fast
#161 opened by mmlingyu - 0
- 1
Swapping between two different AttentionProcessors
#158 opened by Birch-san - 1
Doubts about geglu implementation
#157 opened by capyun - 1
- 1
Error occurs when enabling xformers
#156 opened by wxiwnd - 7
- 4
`RuntimeError: std::bad_alloc` Error occur in other pixel ratio (none-square generation)
#153 opened by Suprhimp - 0
Stable-Diffusion-3 Comming
#151 opened by CallmeZhangChenchen - 4
- 7
Install fatal error C1083: windows11
#145 opened by Nucleon729 - 0
Switching ControlNet and LoRAs
#152 opened by haihp02 - 0
Is stable-fast is available for use in SD WebUi?
#150 opened by bigmover - 2
opening enable_cuda_graph only support two different size image when reference
#141 opened by zhangp365 - 1
- 0
Would stable-fast support (really) old models?
#148 opened by TheodoreGalanos - 0
gcc "-std=99" may need to be declared in code
#147 opened by Fengshawn - 8
- 1
Stable-fast compatibility with lightning models
#146 opened by Davin05 - 6
FP8 support in stable fast
#128 opened by jkrauss82 - 1
Issue of too much recompilation after modification of cross attention processor
#131 opened by xziayro - 1
- 1
- 1
enable_cuda_graph 影响Lora的切换
#143 opened by CallmeZhangChenchen - 1
- 0
Support IP-Adapters
#142 opened by Muawizodux - 3
Changing working device will cause problem
#139 opened by xiyanjoy - 2
Installation failed with an error
#129 opened by cavaniho - 1
切换ControlNets
#137 opened by callmezhangchenchenokay - 0
xformers 25 removes Triton fmha and completely breaks stable-fast, with fix (MemoryEfficientAttentionTritonFwdFlashBwOp, TritonFlashAttentionOp)
#135 opened by tau0-deltav - 0
v1.0.4 release contains the prebuilt binary for 1.0.3
#133 opened by Zxilly - 4
Possible support for stable-cascade
#125 opened by JohnnyRacer - 0
How can I load the whole model from the compiled one instead of loading only the unet to current existing sd model?
#130 opened by quocanh34 - 2
- 1
Tensor Dimension Mismatch
#122 opened by alecyan1993 - 1
any gains to be had on sdxl-turbo?
#123 opened by vibber - 18
Getting This To Work With SD.Next
#119 opened by gohan2091 - 1
Inference Error for with Dynamix Resolution
#126 opened by alecyan1993 - 1
[Feature Request] Support for webui-forge?
#124 opened by banjaminicc - 1
Deepcopy model for further use
#114 opened by alecyan1993 - 0
- 1
Can we use some of the optimization functions on other models unrelated to SD and diffusion ?
#120 opened by oxysoft - 4
Cuda OOM issues
#117 opened by alecyan1993 - 1
get rid of OpenCV related deps, possible?
#113 opened by wingenlit