Tutorial Create Realistic Talking AI Videos Using Infinite Talk with ComfyUI + RunPod Serverless

Infinite Talk Resources

Everything you need for talking AI video generation setup

1. Create Required Accounts

• RunPod – for GPU pods and future serverless functions:
  Referral link: https://runpod.io?ref=ckrxhc11
  Signing up and depositing $10 using this referral may grant you a one-time credit bonus between $5–$500.
• HuggingFace – required for accessing and downloading AI models:
  https://huggingface.co/

2. Fund Your RunPod Account

Ensure you have at least $10 in your RunPod account to spin up GPU pods and receive the bonus credit (if using referral).

Referral link: https://runpod.io?ref=ckrxhc11

3. Create Network Storage

• Go to the "Network Volumes" section in RunPod.
• Create a new volume with at least 70 GB of space.
• This will allow persistent storage between pod sessions.

4. Deploy a Pod with GPU

1. From the RunPod dashboard, click "Deploy Pod".
2. Select a GPU (e.g., RTX A5000 for better performance).
3. Under Templates, choose: Next diffusion-Comfyui sage ateention
4. Attach your network volume during setup.

5. Launch the Environment

• Wait for the pod to fully install (may take several minutes).

6. Upload the Workflow

Upload the Infinite Talk workflow file provided in the download files to your ComfyUI setup.

7. Install Necessary Models

Download and install all required Infinite Talk models for talking video generation.

8. Test the Workflow

Make sure all the model names are updated to match the ones we downloaded.

• Open ComfyUI and load your Infinite Talk workflow
• Verify all model paths point to the correct downloaded files
• Test the workflow with a sample image and audio to ensure it generates talking video correctly

9. Move the Models Folder

Move the models folder from ComfyUI to workspace root.

10. Clean Up Workspace

Remove ComfyUI folder and other unnecessary folders. Keep only the models folder in workspace.

This will leave only the models folder in your workspace, ready for serverless deployment.

11. Terminate the Pod

All files are stored in Network Storage and can be accessed by the serverless endpoint.

• Terminate the current pod to save costs
• All models and files remain accessible via Network Storage

12. Upload to Private GitHub Repository

Now we will start setting up the serverless deployment.

• Upload the given Infinite Talk serverless GitHub folder to Private GitHub Repository
• Create a private repo and push your ComfyUI files, Dockerfile, and snapshot
• Make sure large model files are not tracked unless necessary

13. Deploy as Serverless Endpoint

• Connect your GitHub repo on your RunPod
• Choose the GitHub repo
• Add HuggingFace token for model access

14. Configure Endpoint Settings

After successful deployment, edit the endpoint and add the following:

• Add network storage where you downloaded all the models
• Add new environment variables:

15. Save and Wait

Save the configuration and wait for the deployment to finish.

16. Testing the Endpoint

Two methods available for testing:

• Method 1: Postman
  • Upload the Postman collection file uploadtopostman.postman_collection.postman_collection.json from the download files to Postman
  • Copy the code from wan_api_postman_infinitetalk_wan2.1.json in Postman to send requests
  • Make sure to add your API key and endpoint ID in the Postman request
• Method 2: Custom Web App

Notes

• If you modified ComfyUI, update your Postman and app code accordingly
• Match API structure, input keys, and output format, model names, etc.

Summary Checklist

This tutorial provides everything you need to create your own AI talking video generation app using ComfyUI Infinite Talk on RunPod. All files, code, and templates are available for FREE!