onediff

onediff copied to clipboard

Repeated compilation of the txt2image model under dynamic resolution

3

### Describe the bug #### A clear and concise description of what the bug is. When using input data with different resolution sizes, the program compiles the graphical model multiple...

lijunliangTG

After using onediff to speed up, it slows down in sd-webui

1

### Describe the bug #### A clear and concise description of what the bug is. After using onediff to accelerate, it is slower than before torch ### Your environment ####...

lijunliangTG

### Describe the bug If I use 'nsys profile' to run the script https://github.com/siliconflow/onediff/blob/main/onediff_diffusers_extensions/examples/text_to_image.py, the segmentation fauls happends. ### Your environment ubuntu 20.04，torch==2.0.1+cu118，diffusers==0.27.2, transformers==4.40.1， GPU A10 #### OneDiff git commit...

chenly15

Diffusers IPAdapter bug

21

### Describe the bug #### I have the baseline text_to_image_sdxl_light.py working (using lora checkpoint) and tried adding ipadapter to it. Any ipadapter examples would be great, ideally I'm looking to...

joel-simon

During execution, the script keeps adjusting the size of the output image, resulting in OOM of the graphics card

3

### Describe the bug #### A clear and concise description of what the bug is. Load and execute an image-to-image conversion model based on Stable Diffusion XL, accelerated using OneDiff....

lijunliangTG

Compilation: AttributeError: 'str' object has no attribute 'removeprefix'

2

### Describe the bug I get the following error when compilation starts ``` File "/opt/venv/lib/python3.8/site-packages/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion.py", line 971, in __call__ noise_pred = self.unet( File "/opt/venv/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1511, in _wrapped_call_impl return self._call_impl(*args,...

karlbernard2

### Describe the bug 使用nsys时会在import oneflow as flow处就报错，不使用nsys程序能正常运行，跑的例子是这个 https://github.com/siliconflow/onediff/blob/main/onediff_diffusers_extensions/examples/text_to_image.py ### Your environment ubuntu 20.04，torch==2.0.1+cu118，diffusers==0.27.2, transformers==4.40.1， GPU A10 #### OneDiff git commit id b9d0ae82445ac858206bc12a8fe8dfd765e75d65 #### OneFlow version info version: 0.9.1.dev20240427+cu118 git_commit:...

chenly15

在webui中使用regional-prompter插件，编译模型时会报错。 插件地址：https://github.com/hako-mikan/sd-webui-regional-prompter 插件有2种模式，分别是 Attention 和 Latent 其中Latent模式可以使用onediff加速 Attention模式在加载模型时会报错，错误内容如下 (venv) [root@VM-0-64-centos stable-diffusion-webui]# sh webui.sh -f --no-half-vae --skip-install --listen --port 7863 --disable-safe-unpickle --skip-version-check --controlnet-annotator-models-path /aigc_dev/stable-diffusion-webui/models/ --enable-insecure-extension-access ################################################################ Install script for stable-diffusion +...

jjl001

dev_support_diffusers_ipa

1

## Install: ### step 1: `pip install diffusers==0.27` ### step2. [OneDiff Installation Guide](https://github.com/siliconflow/onediff/blob/main/README_ENTERPRISE.md#install-onediff-enterprise) ### step3. [OneDiffx Installation Guide](https://github.com/siliconflow/onediff/tree/main/onediff_diffusers_extensions#install-and-setup) ## Usage: script_00.py ```python """ ### Install ## Prepare environment You need...

ccssu

Dev ipa+isntantid

3

fix: https://github.com/siliconflow/onediff/issues/795 ## Doc - [ ] join ci - [x] [Accelerating ComfyUI_InstantID with OneDiff](https://github.com/siliconflow/onediff/tree/dev_ipa%2Bisntantid/onediff_comfy_nodes/modules/oneflow/hijack_comfyui_instantid#readme) - [x] [Accelerating ComfyUI_IPAdapter_plus with OneDiff](https://github.com/siliconflow/onediff/blob/3d5010e7be267c2f1cd10e413285f1e07d7b74a8/onediff_comfy_nodes/modules/oneflow/hijack_ipadapter_plus/README.md) - [ ] [Accelerating ComfyUI_IPAdapter_plus + ComfyUI_InstantID with OneDiff]()...

ccssu