Releases: YanWenKun/ComfyUI-Windows-Portable
Releases · YanWenKun/ComfyUI-Windows-Portable
Wheel Files for ComfyUI-3D-Pack 2024/10/27
2024/10/25 Without 3D-Pack
Note | 提示
- With the main components using the latest versions, this release may have better performance for new models like FLUX and SD 3.5.
- 本次发布版本追新，意在 Flux 与 SD 3.5 等新模型的性能。
- This release doesn't include:
- 本次发布不含：
- 3D-Pack
- InstantID
- ReActor
- AnimateDiff
- AnimateAnyone
- FaceAnalysis
- WAS NS
- MTB Nodes
Hint | 使用技巧
- If you're using 40 series / Ada GPU or newer, you can try adding
--fast to the startup arguments in run_nvidia_gpu.bat.
- 如果你在用 40 系或更新 GPU，可以尝试开启实验性参数
--fast，编辑 run_nvidia_gpu.bat 添加至启动行末尾即可。
- Source: comfyanonymous/ComfyUI@9953f22
- If you want to turn-off xFormers, add
--use-pytorch-cross-attention to the startup arguments in run_nvidia_gpu.bat.
- 如果你不想启用 xFormers，编辑
run_nvidia_gpu.bat 并在启动参数里添加 --use-pytorch-cross-attention。
Versions | 版本信息
- Python 3.12.7
- CUDA 12.4
- PyTorch 2.5.0
- xFormers 0.0.28.post2
2024/10/18
| How to use: |
如何使用 |
1. Download all .7z.00X files |
1. 下载所有 .7z.00X 文件 |
2. Open the .001 file and extract |
2. 打开第一个文件（.001）并解压 |
3. Edit run_nvidia_gpu.bat to fit your GPU |
3. 编辑 run_nvidia_gpu.bat 以适配你的 GPU |
4. Run run_nvidia_gpu.bat to start ComfyUI |
4. 运行 run_nvidia_gpu.bat 启动 ComfyUI |
| Note: You can install custom nodes via "ComfyUI-Manager" |
提示：使用“管理器”可以安装自定义节点 |
Changes | 变动
- Add script
reinstall-deps-for-3d.bat that build and re-install dependencies for Comfy3D.
- 添加脚本
重新编译安装3D-Pack依赖项.bat 。
Versions | 版本信息
- Python 3.12.7
- CUDA 12.1
- PyTorch 2.4.0
- xFormers 0.0.27.post2
2024/10/02
| How to use: |
如何使用 |
1. Download all .7z.00X files |
1. 下载所有 .7z.00X 文件 |
2. Open the .001 file and extract |
2. 打开第一个文件（.001）并解压 |
3. Run run_nvidia_gpu.bat to start ComfyUI |
3. 运行 run_nvidia_gpu.bat 启动 ComfyUI |
| Note: You can install custom nodes via "ComfyUI-Manager" |
提示：使用“管理器”可以安装自定义节点 |
Changes | 变动
- Upgrade to Python 3.12
- 升级到 Python 3.12
Versions | 版本信息
- Python 3.12.6
- CUDA 12.1
- PyTorch 2.4.0
- xFormers 0.0.27.post2
2024/10/01
| How to use: |
如何使用 |
1. Download all .7z.00X files |
1. 下载所有 .7z.00X 文件 |
2. Open the .001 file and extract |
2. 打开第一个文件（.001）并解压 |
3. Run run_nvidia_gpu.bat to start ComfyUI |
3. 运行 run_nvidia_gpu.bat 启动 ComfyUI |
| Note: You can install custom nodes via "ComfyUI-Manager" |
提示：使用“管理器”可以安装自定义节点 |
Note | 提示
- This is the final release based on Python 3.11. Future releases will be based on Python 3.12.
- 该版为最后一次基于 Python 3.11 打包，后续发布将基于 Python 3.12。
- Only few subtle changes have made. You don't need to update to this one if you're already using 2024/09/29 release.
- 仅包含少量变更，已使用 2024/09/29 版的用户无需更新。
Changes | 变动
- Forcing LF line ending for git.
- In
run_nvidia_gpu.bat:
- Better pycache path.
- Remove unnecessary quotes in PIP mirror and HF mirror URL.
- 强制 git 使用 LF 行尾
- 启动脚本
run_nvidia_gpu.bat 变动：
- pycache 使用绝对路径
- 去掉 PIP 和 HF 镜像 URL 中多余的引号
Versions | 版本信息
- Python 3.11.9
- CUDA 12.1
- PyTorch 2.4.0
- xFormers 0.0.27.post2
2024/09/29
| How to use: |
如何使用 |
1. Download all .7z.00X files |
1. 下载所有 .7z.00X 文件 |
2. Open the .001 file and extract |
2. 打开第一个文件（.001）并解压 |
3. Run run_nvidia_gpu.bat to start ComfyUI |
3. 运行 run_nvidia_gpu.bat 启动 ComfyUI |
| Note: You can install custom nodes via "ComfyUI-Manager" |
提示：使用“管理器”可以安装自定义节点 |
Changes | 变动
- Upgrade to PyTorch 2.4.0
- Changes in
run_nvidia_gpu.bat:
- Add mirror site settings for PIP & HuggingFace Hub
- Will try to copy
u2net.onnx to user's home path, to avoid extra downloading at first start
- 升级到 PyTorch 2.4.0
- 启动脚本
run_nvidia_gpu.bat 增加功能
- 可以为 PIP 与 HuggingFace Hub 设置镜像站点，优化国内下载速度。需手动编辑启用
- 会尝试复制
u2net.onnx 到用户主目录，以避免初次启动时下载
Versions | 版本信息
- Python 3.11.9
- CUDA 12.1
- PyTorch 2.4.0
- xFormers 0.0.27.post2
2024/09/29 Without 3D-Pack
Note | 提示
- This release is for Flux and performance, using some newest components.
- 本次发布意在 Flux 与高性能，主要组件使用最新版本。
- This release doesn't include:
- 本次发布不含：
- Comfy3D
- InstantID
- ReActor
- AnimateDiff
- AnimateAnyone
- FaceAnalysis
- WAS NS
- MTB Nodes
Hint | 使用技巧
- If you're using 40 series / Ada GPU or newer, you can try adding
--fast to the startup arguments in run_nvidia_gpu.bat.
- 如果你在用 40 系或更新 GPU，可以尝试开启实验性参数
--fast，编辑 run_nvidia_gpu.bat 添加至启动行末尾即可。
- Source: comfyanonymous/ComfyUI@9953f22
Versions | 版本信息
- Python 3.12.6
- CUDA 12.4
- PyTorch 2.4.1
- xFormers 0.0.28.post1
2024/09/10
| How to use: |
如何使用 |
1. Download all .7z.00X files |
1. 下载所有 .7z.00X 文件 |
2. Open the .001 file and extract |
2. 打开第一个文件（.001）并解压 |
3. Run run_nvidia_gpu.bat to start ComfyUI |
3. 运行 run_nvidia_gpu.bat 启动 ComfyUI |
| Note: You can install custom nodes via "ComfyUI-Manager" |
提示：使用“管理器”可以安装自定义节点 |
Changes | 变动
- Add environment variable
TORCH_CUDA_ARCH_LIST to start script, fixing Comfy3D errors.
- Recommend edit
run_nvidia_gpu.bat before start, it has instruction comments.
- 添加环境变量
TORCH_CUDA_ARCH_LIST 到启动脚本，解决 Comfy3D 错误。
- 推荐启动前先编辑一下
run_nvidia_gpu.bat，内有说明。
Versions | 版本信息
- Python 3.11.9
- CUDA 12.1
- PyTorch 2.3.0
- xFormers 0.0.26.post1
2024/09/08
| How to use: |
如何使用 |
1. Download all .7z.00X files |
1. 下载所有 .7z.00X 文件 |
2. Open the .001 file and extract |
2. 打开第一个文件（.001）并解压 |
3. Run run_nvidia_gpu.bat to start ComfyUI |
3. 运行 run_nvidia_gpu.bat 启动 ComfyUI |
| Note: You can install custom nodes via "ComfyUI-Manager" |
提示：使用“管理器”可以安装自定义节点 |
Changes | 变动
- A new start script that may help preventing PATH problems.
- 新的启动脚本，有望避免 PATH 环境变量的问题。
Versions | 版本信息
- Python 3.11.9
- CUDA 12.1
- PyTorch 2.3.0
- xFormers 0.0.26.post1
2024/09/08 Without 3D-Pack
Note | 提示
- This release is focus on Flux, using some of the newest components.
- 本次发布意在 Flux，主要组件使用最新版本，请留意版本信息。
- This release doesn't include:
- 本次发布不含：
- Comfy3D
- InstantID
- ReActor
- AnimateDiff
- AnimateAnyone
- FaceAnalysis
- WAS NS
Hint | 小技巧
- If you're using 40 series / Ada GPU or newer, you can try adding
--fast to the startup arguments in run_nvidia_gpu.bat.
- 如果你在用 40 系或更新 GPU，可以尝试开启实验性参数
--fast，编辑 run_nvidia_gpu.bat 添加至启动行末尾即可。
- Ref: comfyanonymous/ComfyUI@9953f22
Versions | 版本信息
- Python 3.12.6
- CUDA 12.4
- PyTorch 2.4.1
- xFormers NOT installed
