ComfyUI is a powerful, open-source image generation tool. A core feature of ComfyUI is its support for a wide variety of models, which determine the content and style of the generated images. This guide introduces several methods for downloading ComfyUI models, allowing you to efficiently obtain the models you need.
1. Model Square: One-stop download for curated models
Model Square simplifies the process of obtaining models and overcomes the challenges of downloading them over international networks. Model Square aggregates popular ComfyUI models to help you easily find and download what you need.
Download steps:
Log on to the console: Function AI console. In the navigation pane on the left, click Projects. Then, in the project list, click the target image generation project.
Access Model Square: Click the Model Square tab on the image generation project details page.
Browse and select: Model Square displays a clear list of models. You can filter and browse by model type and source address to find the model that best suits your needs.
Use the filter bar to display models of a specific category, such as all Checkpoint models.
One-click download: To download a model, click the
icon in its upper-right corner. In the Add to My Models dialog box, confirm the model file and storage path, and then click OK.You can also click the model to view its details. Click the Download Model button. In the Add to My Models dialog box, confirm the model file and storage path, and then click OK.
After the download begins, a task list automatically appears to show the download progress.
Once the download is complete, go to the File Management tab and find the downloaded model in the
modelsdirectory. You can now use this model in your ComfyUI project.
2. File management: Flexibly upload local or cloud models
If you have model files on your local computer or in Object Storage Service (OSS), you can use the File Management feature to upload them to ComfyUI.
2.1 Upload a model from OSS
Log on to the console: Log on to the Function AI console. In the navigation pane on the left, click Projects. On the Projects page, click the target image generation project.
Go to File Management: Click the File Management tab on the image generation project details page.
Go to the model directory: Navigate to the subdirectory where you want to upload the model, such as the
/models/lorasdirectory.Import a model from OSS: In the model directory, click Import from OSS in the upper-right corner. In the dialog box that appears, set Bucket to your OSS bucket. Select the checkbox for the model file to upload, and then click Import. The system pulls the model file from OSS and imports it to the selected destination path.
2.2 Upload a model from a local computer
You can upload model files directly from your local computer using the File Management feature.
Log on to the console: Log on to the Function AI console. In the navigation pane on the left, click Projects. On the Projects page, click the target image generation project.
Go to File Management: Click the File Management tab on the image generation project details page.
Local upload: Go to the destination subdirectory, such as
/models/loras. Then, click Local Upload in the upper-right corner.Select the model file: The system opens a file selector. Browse your local file system to find and select the ComfyUI model file that you want to upload.
Confirm the upload: Click the Upload button to upload the model file to your ComfyUI environment.
3. Use the Hugging Face and ModelScope command-line tools
If you are familiar with the command line or need to download the latest models, the most efficient method is to use the command-line tools for Hugging Face Hub and ModelScope. Both platforms are leading communities for open-source AI models.
3.1 Download Hugging Face models using huggingface_cli
Hugging Face Hub hosts tens of thousands of pre-trained models. The official command-line interface (CLI), huggingface_cli, lets you download these models easily.
Prerequisites
Ensure that the huggingface_hub Python library is installed in your environment. If not, on the image generation project details page, click . On the instance page, run the following command to install the huggingface_hub Python library.
pip install huggingface_hubDownload command
Use the following command to download a model. You can find the full ID for each model on the Hugging Face Hub website.
For more information about downloading models with huggingface-cli, see the Hugging Face documentation.
huggingface-cli download <model_id> --local-dir <local_directory><model_id>: The full identifier of the model that you want to download from Hugging Face Hub. For example, the identifier for https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/ae.safetensors isblack-forest-labs/FLUX.1-dev ae.safetensors.--local-dir <local_directory>: The path on your local file system where the model will be saved after the download. Make sure that the path exists and that you have write permission.
Download example
Download the ae.safetensors model and save it to the ./models/vae directory:
huggingface-cli download black-forest-labs/FLUX.1-dev ae.safetensors --local-dir ~/comfyui/models/vae/
3.2 Download models using ModelScope
ModelScope is an open-source AI model community led by Alibaba DAMO Academy. It provides a rich collection of Chinese-language and multimodal models.
Prerequisites
The modelscope Python library must be installed in your environment. If it is not installed, on the image generation project details page, click . On the instance page, run the following command to install the library.
pip install modelscopeDownload command
Use the following command to download a model. For more information about downloading models with ModelScope, see Download models.
modelscope download <model_id> --local_dir <local_directory><model_id>: The full identifier of the model that you want to download from ModelScope. For example, the identifier for https://www.modelscope.cn/models/black-forest-labs/FLUX.1-dev/file/view/master/ae.safetensors isblack-forest-labs/FLUX.1-dev ae.safetensors.--local_dir <local_directory>: The path on your local file system where the model will be saved after the download. Make sure that the path exists and that you have write permission.
Download example
Download the ae.safetensors model and save it to the ./models/vae directory:
modelscope download black-forest-labs/FLUX.1-dev ae.safetensors --local_dir '~/comfyui/models/vae'More information
AI technology develops quickly, and new models emerge constantly. Follow the updates in Model Square to obtain the latest models and expand the creative boundaries of ComfyUI.
If you have any questions or need help while you explore the world of ComfyUI models, you can read our help documents or contact our technical support team. Happy creating!