siliconflow/onediff
OneDiff: An out-of-the-box acceleration library for diffusion models.
PythonApache-2.0
Issues
- 2
Cannot import /comfyui/custom_nodes/onediff_comfy_nodes module for custom nodes: No module named 'onediff.infer_compiler'`
#920 opened by alboto - 0
Diffusers StableDiffusionInpaintPipeline OOM
#922 opened by Someant - 2
- 1
是否支持Nvidia T4或者L4 GPU?
#906 opened by dawei03896 - 12
- 1
可以加速 ipadapter 吗?
#918 opened by zhangyongyu - 8
TypeError: scaled_dot_product_attention(): argument 'query' (position 1) must be Tensor, not Tensor
#890 opened by BradVidler - 1
- 0
- 25
Diffusers IPAdapter bug
#839 opened by joel-simon - 3
Compiled ControlNetModel ignores controlnet_conditioning_scale and control_guidance_end
#899 opened by jamarju - 0
- 0
- 1
- 1
failed to start webui with onediff
#889 opened by XRFate - 0
- 4
- 0
"LayerNormKernelImpl" not implemented for 'Half'
#903 opened by liuchenbaidu - 1
有计划支持ComfyUI的PuLID吗?
#901 opened by huan085128 - 0
Doc Improve: To use onediff in docker pytorch/pytorch:2.2.2-cuda12.1-cudnn8-runtime you need add LD_LIBRARY_PATH to load libcudnn_cnn_infer.so.8
#896 opened by ganlvtech - 2
Found a small problem with the calculation
#893 opened by songh11 - 2
Source code of oneflow
#861 opened by rfan-debug - 0
- 3
No module named 'onediff.infer_compiler.register'
#884 opened by Ralphhtt - 2
- 0
oneflow compile error
#880 opened by nono-Sang - 1
Support for custom attention processor?
#873 opened by Davin05 - 0
DeepCache error and recompilation in ComfyUI
#857 opened by nono-Sang - 5
- 0
Is it possible to manually set recompile to false?
#870 opened by Davin05 - 1
请问一下,我现在想要加速sdxl版本的图生图,输入图像的分辨率是任意的,这样情况可以用oneflow加速吗?
#866 opened by cjt222 - 0
ComfyUI_IPAdapter_plus with OneDiff支持再加lora吗
#859 opened by xiecon - 3
comfyui切换模型报错
#865 opened by r970707452c - 2
webui 生成非32倍分辨率的图片会报错
#832 opened by huan085128 - 0
- 2
Compilation: AttributeError: 'str' object has no attribute 'removeprefix'
#831 opened by karlbernard2 - 3
During execution, the script keeps adjusting the size of the output image, resulting in OOM of the graphics card
#848 opened by lijunliangTG - 3
- 0
runtime and compiletime cudu is not match
#852 opened by SmallStom - 0
Can not use 'nsys profile' for onediff diffusers
#850 opened by chenly15 - 4
can't install in comfyui
#827 opened by Uniky - 0
oneflow优化和nsys同时使用时出错
#845 opened by chenly15 - 0
webui中使用regional-prompter插件时编译模型报错
#844 opened by jjl001 - 5
让 ComfyUI Manager 安装 Onediff 更丝滑
#838 opened by nono-Sang - 3
Compilation optimization issue
#835 opened by chuck-ma - 2
Can OneDiff use for webui-forge?
#829 opened by akk-123 - 0
image-to-image sdxl light
#841 opened by Muawiz-Umar - 8
转换sam的ImageEncoderViT模型失败
#824 opened by zhangvia - 0
[New algo for optimization]Align Your Steps: Optimizing Sampling Schedules in Diffusion Models
#840 opened by strint - 0
Could not find a version that satisfies the requirement onediff-quant (from versions: none)
#828 opened by ziQzav