3D Model Generation from Images using Hunyuan3D via Replicate - n8n Workflow

3D Artists and Designers looking to automate asset creation.
Developers building automated content pipelines.
Users seeking efficient n8n templates for AI model interaction.
Anyone needing a reliable n8n workflow for long-running, asynchronous API tasks.

Generating complex 3D assets using AI models like Hunyuan3D is typically an asynchronous process. This means the API call starts the job but doesn't immediately return the final result; you must continuously check (poll) its status. This dedicated n8n workflow solves this complexity by providing a robust polling mechanism built around the standard n8n node functionality.

This specific n8n workflow utilizes an HTTP Request n8n node to initiate the generation on Replicate and then uses a structured loop (Wait and If n8n nodes) to repeatedly check the job status until the 3D model generation is complete. It transforms a lengthy development challenge into a reusable, efficient n8n template, ensuring reliable output retrieval.

This comprehensive n8n workflow is initiated by a manual n8n trigger, designed for easy testing and integration.

1. Initialization: The workflow starts with the 'On clicking 'execute'' n8n trigger. It immediately proceeds to the 'Set API Key' n8n node, which securely stores your Replicate API key for subsequent requests.


2. Job Creation: The 'Create Prediction' HTTP Request n8n node sends a POST request to Replicate, specifying the ndreca/hunyuan3d-2-test model version and defining input parameters like the source image URL, seed, and optimization settings.


3. Polling Setup: The 'Extract Prediction ID' Code n8n node processes the initial response, which contains the job ID and status. It extracts the necessary data points, including the unique ID and the dedicated polling URL.


4. Waiting Period: The 'Wait' n8n node pauses the n8n workflow for 2 seconds to prevent excessive API calls while the model processes.


5. Status Check: The 'Check Prediction Status' HTTP Request n8n node uses the previously extracted URL to query Replicate for the current status of the job.


6. Looping Logic: The 'Check If Complete' If n8n node evaluates the returned status. If the status is 'succeeded', the n8n workflow proceeds to finalize the result. If the job is still running (e.g., 'processing' or 'starting'), the workflow loops back to the 'Wait' n8n node, restarting the polling cycle until the model is finished.


7. Final Output: Once complete, the 'Process Result' Code n8n node extracts the final output (the generated 3D model file/URL) and relevant metadata, delivering the final structured data to the end of the n8n workflow.

To set up this powerful n8n workflow template, follow these steps:

1. Import: Copy the provided JSON data and import it directly into your n8n instance using the 'New' > 'Import from JSON' feature.


2. API Key Configuration: Locate the 'Set API Key' n8n node. Update the 'replicateapikey' assignment with your actual Replicate API key. Ensure the value replaces 'YOURREPLICATEAPI_KEY'.


3. Input Configuration: Navigate to the 'Create Prediction' HTTP Request n8n node. Within the JSON body, update the placeholder "image": "https://example.com/input.other" with the URL of the image you wish to convert into a 3D model.


4. Credentials: This n8n workflow uses Generic Credentials defined within the HTTP Request n8n node for authentication. No separate n8n credential setup is needed, as the Bearer token is constructed dynamically using the API key set in the previous n8n node.


5. Execution: Activate the n8n workflow and run it using the 'On clicking 'execute'' n8n trigger.

On clicking 'execute' (Manual Trigger n8n trigger): Serves as the starting point for manual testing and execution of the n8n workflow.
Set API Key (Set n8n node): Essential for secure operation. It sets the replicateapikey variable, which is dynamically passed as a Bearer token in subsequent HTTP Request n8n nodes.
Create Prediction (HTTP Request n8n node): Initiates the AI generation job on Replicate. Key configuration includes using a POST method, sending a JSON body with input image details, and setting the Authorization header using the API key variable.
Extract Prediction ID (Code n8n node): A custom scripting n8n node that extracts the unique ID and constructs the necessary polling URL (predictionUrl) for tracking the asynchronous job status.
Wait (Wait n8n node): Implements a necessary delay (2 seconds) to manage the polling frequency and adhere to API rate limits during the asynchronous execution of this n8n workflow.
Check Prediction Status (HTTP Request n8n node): The core polling n8n node. It repeatedly fetches the status of the job using the dynamically created prediction URL.
Check If Complete (If n8n node): Controls the loop structure. It checks if $json.status is 'succeeded'. If not, the n8n workflow flow returns to the 'Wait' node.
Process Result (Code n8n node): The final transformation n8n node, standardizing the successful output data, extracting the 3D model URL, and providing execution metrics.