Google Veo 3 Text-to-Video Generation via Replicate API - n8n Workflow

• Content Creators needing automated video generation.


• Marketing teams looking to rapidly prototype video ads using AI.


• Developers seeking robust n8n templates for integrating advanced AI models.


• Users who need a reliable n8n workflow to handle asynchronous API tasks like long-running video creation.

Generating high-quality video content from text prompts often involves long processing times and requires continuous API polling, which can be complicated to set up manually. This comprehensive n8n workflow solves this by seamlessly managing the entire lifecycle: from initializing the video generation request (using the google/veo-3 model) to automatically polling the status until completion. This reusable n8n template ensures you get the final video URL without manual intervention. By using the core HTTP Request n8n node, this solution remains flexible and powerful, making it an essential addition to your collection of n8n templates.

This automation starts with a manual n8n trigger, designed for testing or integration into a larger schedule.

1. Initialization: The n8n workflow first uses a Set n8n node to securely define the Replicate API key.


2. Creation Request: An HTTP Request n8n node sends a POST call to Replicate's API, initiating the video generation process for the specified prompt using the Google Veo 3 model.


3. ID Extraction: A Code n8n node immediately parses the response to extract the unique Prediction ID and constructs the necessary polling URL.


4. Asynchronous Polling: The workflow enters a crucial polling loop. It begins with a Wait n8n node, pausing for 2 seconds.


5. Status Check: A subsequent HTTP Request n8n node checks the prediction status using the extracted URL and API key.


6. Loop Control: An If n8n node evaluates the status. If the status is 'succeeded', the process moves forward. If not, the n8n workflow loops back to the Wait n8n node, effectively polling until the video is ready.


7. Final Output: Once complete, a final Code n8n node processes the result, extracting the final generated video URL and relevant metadata, concluding the n8n workflow execution.

To deploy this powerful n8n workflow, follow these steps:

1. Import: Copy the provided JSON code and import it into your n8n instance via the 'Workflows' section.


2. API Key Setup: Locate the 'Set API Key' n8n node. Replace the placeholder value YOURREPLICATEAPIKEY with your actual Replicate API key.


3. Prompt Configuration: Locate the 'Create Prediction' HTTP Request n8n node. In the JSON Body, update the prompt value to the desired text description for your video.


4. Credentials: Ensure you have configured any generic HTTP Header Authentication correctly, linking the authorization header to the replicateapi_key set in the previous n8n node.


5. Execution: Run the n8n workflow manually using the 'On clicking 'execute'' n8n trigger to generate your first video.

On clicking 'execute' (n8n trigger): A manual n8n trigger used to initiate the workflow for testing and immediate use.
Set API Key (Set n8n node): Essential for security, this node stores the Replicate API key as an expression variable (replicateapikey) used by subsequent HTTP Request n8n nodes for authentication.
Create Prediction (HTTP Request n8n node): Sends the initial request to the Replicate service. Key configuration includes the POST method, setting the Authorization header using the extracted API key, and providing the video generation prompt in the JSON body.
Extract Prediction ID (Code n8n node): Uses JavaScript to parse the API response, extracting the unique ID required for polling and constructing the status check URL. This is a vital part of the n8n workflow's logic.
Wait (Wait n8n node): Implements a necessary 2-second delay in the polling loop to prevent hitting API rate limits while waiting for the video generation to complete.
Check Prediction Status (HTTP Request n8n node): The core polling n8n node. It periodically checks the status of the long-running video generation task until the result is available.
Check If Complete (If n8n node): Controls the polling loop. It checks if the status is 'succeeded'. If not, the n8n workflow cycles back to the Wait node.
Process Result (Code n8n node): If generation is successful, this n8n node structures the final output, specifically isolating the generated video_url for downstream use.