Text Translation and Generation with Replicate Seed-X-PPO Model - n8n Workflow

Markdown-formatted text:
Developers and engineers needing reliable asynchronous job handling in their automation.
Users who want to leverage powerful, specialized AI models like lucataco/seed-x-ppo within an n8n environment.

• Teams looking for robust, reusable n8n templates for handling AI tasks that require a long processing time.

Many advanced AI models hosted on platforms like Replicate require significant processing time, meaning the initial request only kicks off a job asynchronously. This presents a challenge for linear automation platforms.

This specific n8n workflow addresses this critical issue by implementing a resilient polling loop using standard n8n node logic. This powerful n8n workflow starts the job, extracts the prediction ID, and then repeatedly polls the Replicate API until the job status changes to 'succeeded'. This ensures that you reliably capture the final text output, making this n8n template invaluable for high-latency AI operations. The automation setup is clean, requiring only your Replicate API key to get started.

This n8n workflow operates in a distinct sequence designed for asynchronous API interaction:

1. Start the Process: The execution begins with the Manual n8n trigger node, allowing immediate testing and manual initiation.


2. Authentication: The 'Set API Key' n8n node securely stores the Replicate API key needed for subsequent HTTP requests.


3. Initiate AI Job: The 'Create Prediction' HTTP Request n8n node sends the text input and target language to Replicate, starting the translation/generation job and receiving an initial job ID.


4. Prepare for Polling: The 'Extract Prediction ID' Code n8n node parses the initial response to pull out the unique ID and construct the URL required for checking the status of that specific job.


5. Polling Loop: The workflow enters a polling sequence. It first hits the 'Wait' n8n node for 2 seconds to prevent excessive API calls.


6. Check Status: The 'Check Prediction Status' HTTP Request n8n node retrieves the current job status using the prediction URL derived in Step 4.


7. Conditional Check: The 'Check If Complete' If n8n node evaluates the status. If the status is 'succeeded', it proceeds to 'Process Result'. If the status is anything else (e.g., 'processing'), the n8n workflow loops back to the 'Wait' n8n node to re-poll.


8. Final Output: Once successful, the 'Process Result' Code n8n node extracts the generated output text and associated metadata for final consumption.

Markdown-formatted text:

1. Importing: Copy the provided JSON data and navigate to your self-hosted or cloud n8n instance. Use the 'New' menu -> 'Import from JSON' to load this n8n template.


2. API Key Setup: Locate the 'Set API Key' n8n node. Replace the placeholder value YOURREPLICATEAPIKEY with your actual key from the Replicate platform. This step is crucial for authenticating all subsequent requests in the n8n workflow.


3. Input Configuration: Open the 'Create Prediction' HTTP Request n8n node. Within the JSON Body parameter, customize the text you wish to translate/generate and the desired targetlanguage.


4. Testing: Click the 'On clicking 'execute'' n8n trigger node and click 'Execute Workflow' to run the entire n8n workflow and test the polling logic.

Markdown-formatted text:

On clicking 'execute' (Manual Trigger): This serves as the initial n8n trigger, allowing for immediate manual execution and testing of the n8n workflow.
Set API Key (Set n8n node): Functionally, this node sets the required replicateapikey as an item variable, making it accessible for bearer token authentication in later HTTP Request n8n nodes.
Create Prediction (HTTP Request n8n node): Responsible for initiating the AI job on Replicate. It targets the v1/predictions endpoint and uses the lucataco/seed-x-ppo model version (bd6fdc73...).
Extract Prediction ID (Code n8n node): A core logic n8n node that uses JavaScript to dynamically parse the API response, extracting the unique job id and preparing the full predictionUrl required for polling.
Wait (Wait n8n node): Crucial for rate limiting the polling process. It pauses the n8n workflow execution for 2 seconds before the next status check.
Check Prediction Status (HTTP Request n8n node): Performs the status check using the dynamically generated predictionUrl and the stored API key.
Check If Complete (If n8n node): The flow control n8n node that directs the automation. It checks if the $json.status equals 'succeeded'. If false, the flow loops back; if true, it proceeds to output processing.
Process Result (Code n8n node): Cleans and formats the final successful response, extracting the core output text and important metrics into a readable structure.