chengzeyi/stable-fast

Issues

• The content of the image is different from the original

  #112 opened 10 months ago
  1

• Save compiled model and reuse?

  #111 opened 10 months ago
  3

• Support for DeepCache

  #110 opened 10 months ago
  2

• Error when running on A100 Replicate

  #109 opened 10 months ago
  3

• Errors when running on A100 on Replicate

  #108 opened 10 months ago
  0

• torch.jit encounters xformers::efficient_attention_forward_cutlass and throws a RuntimeError

  #107 opened 10 months ago
  2

• compile_unet error for `StableDiffusionLatentUpscalePipeline`

  #106 opened 10 months ago
  6

• help me

  #105 opened a year ago
  1

• Dynamically Switch LoRA error

  #104 opened a year ago
  2

• GPU memory usage is higher than before use

  #103 opened a year ago
  1

• Segmentation fault instead of CUDA OOM for SVD with higher batch size

  #100 opened a year ago
  1

• How to use the inpainting pipeline?

  #99 opened a year ago
  1

• How to avoid the long time required for the first warm up after compile the model

  #98 opened a year ago
  2

• Any recommended config for MarigoldPipeline?

  #96 opened a year ago
  1

• Can Dynamically Switch LoRA support text_encoder?

  #95 opened a year ago
  1

• Is stable_fast faster or slower than Huggingface fast_diffusion?

  #93 opened 10 months ago
  2

• stable-fast version 1.0.0 does not support Windows system？

  #92 opened a year ago
  6

• Can't load/unload lora dynamically

  #91 opened a year ago
  2

• Please build release wheels for arch 8.6, 8.9 and arch 9.0

  #89 opened a year ago
  0

• Perf regression on A100 in v1.0.0+torch212+cu121+xformers0.23.post1 v.s. 0.0.13+torch2.0.0+cu121+xformers0.22patch7

  #88 opened a year ago
  2

• support animatediff

  #87 opened a year ago
  9

• Build from source failed

  #86 opened a year ago
  7

• Enable cuda graph error in `StableDiffusionControlNetReferencePipeline`

  #84 opened a year ago
  6

• How to do Ensemble of Expert Denoisers w/ refiner model

  #83 opened a year ago
  3

• Is it possible to optimize ControlNet while retaining dynamic strength?

  #82 opened a year ago
  1

• Ip_adapter not work with xformers enabled

  #81 opened a year ago
  14

• Is this package compatible with Lycoris LoRAs

  #78 opened a year ago
  3

• RuntimeError: no valid convolution algorithms available in CuDNN

  #77 opened a year ago
  7

• SDXL Swap lora Issue

  #75 opened a year ago
  6

• Potential improvements to stable-fast

  #73 opened a year ago
  3

• Trying to get it working on Windows

  #72 opened a year ago
  2

• Does xformers still matter?

  #71 opened a year ago
  2

• Cant get it runnig... anyone can help please? RTX 3090

  #70 opened a year ago
  5

• [composability] stable-fast + sd-turbo device mismatch

  #69 opened a year ago
  2

• Stable Video Optimizations?

  #67 opened a year ago
  3

• setup.py should not contain import of torch

  #66 opened a year ago
  4

• Does it support LCM Lora? The generated images are very poor

  #65 opened a year ago
  3

• SDXL inference speed up compare?

  #63 opened a year ago
  3

• Fails to compile model in docker, missing installs?

  #62 opened a year ago
  2

• Update torch.compile benchmark on A100 40GB SDv1.5 for torch nightly

  #60 opened a year ago
  10

• What's the advance compared to TensorRT?

  #59 opened a year ago
  3

• exception when using deterministic generation with enable_cuda_graph = True

  #58 opened a year ago
  1

• Found an unsupported argument type in the JIT tracer: at::Generator.

  #56 opened a year ago
  8

• 'torch' module not found during 'stable-fast' installation

  #55 opened a year ago
  9

• Black image with relelases post1-post3 and nightly

  #53 opened a year ago
  5

• Compatible with comfy ui?

  #52 opened a year ago
  1

• RuntimeError: _Map_base::at`

  #49 opened a year ago
  11

• 加微信聊一下

  #42 opened a year ago
  1

• Openai triton conv kernel can't run correctly

  #41 opened a year ago
  3

• Tiny AutoEncoder support

  #40 opened a year ago
  0