Generate 3D Models & Textures from Images with Hunyuan3D AI

This workflow connects n8n → Replicate API to generate 3D-like outputs using the ndreca/hunyuan3d-2.1-test model. It handles everything: sending the request, waiting for processing, checking status, and returning results.

⚡ Section 1: Trigger & Setup

⚙️ Nodes

1️⃣ On Clicking “Execute”

• What it does: Starts the workflow manually in n8n.
• Why it’s useful: Great for testing or one-off runs before automation.

2️⃣ Set API Key

• What it does: Stores your Replicate API Key.
• Why it’s useful: Keeps authentication secure and reusable across HTTP nodes.

💡 Beginner Benefit

• No coding needed — just paste your API key once.
• Easy to test: press Execute, and you’re live.

🤖 Section 2: Send Job to Replicate

⚙️ Nodes

3️⃣ Create Prediction (HTTP Request)

• What it does: Sends a POST request to Replicate’s API with:

  • Model version (70d0d816...ae75f)
  • Input image URL
  • Parameters like steps, seed, generate_texture, remove_background

• Why it’s useful: This kicks off the AI generation job on Replicate’s servers.

4️⃣ Extract Prediction ID (Code)

• What it does: Grabs the prediction ID from the API response and builds a status-check URL.
• Why it’s useful: Every job has a unique ID — this lets us track progress later.

💡 Beginner Benefit

• You don’t need to worry about JSON parsing — the workflow extracts the ID automatically.
• Everything is reusable if you run multiple generations.

⏳ Section 3: Poll Until Complete

⚙️ Nodes

5️⃣ Wait (2s)

• What it does: Pauses for 2 seconds before checking the job status.
• Why it’s useful: Prevents spamming the API with too many requests.

6️⃣ Check Prediction Status (HTTP Request)

• What it does: GET request to see if the job is finished.

7️⃣ Check If Complete (IF Node)

• What it does:

  • If status = succeeded → process results.
  • If not → loops back to Wait and checks again.

💡 Beginner Benefit

• Handles waiting logic for you — no manual refreshing needed.
• Keeps looping until the AI job is really done.

📦 Section 4: Process the Result

⚙️ Nodes

8️⃣ Process Result (Code)

• What it does: Extracts:

  • status
  • output (final generated file/URL)
  • metrics (performance stats)
  • Timestamps (created_at, completed_at)
  • Model info

• Why it’s useful: Packages the response neatly for storage, email, or sending elsewhere.

💡 Beginner Benefit

• Get clean, structured data ready for saving or sending.
• Can be extended easily: push output to Google Drive, Notion, or Slack.

📊 Workflow Overview

🎯 Overall Benefits

✅ Fully automates Replicate model runs
✅ Handles waiting, retries, and completion checks
✅ Clean final output with status + metrics
✅ Beginner-friendly — just add API key + input image
✅ Extensible: connect results to Google Sheets, Gmail, Slack, or databases

✨ In short: This is a no-code AI image-to-3D content generator powered by Replicate and automated by n8n.