🎬 Generate videos from images with Wan 2.2 I2V A14B AI model
⚡ 3,733 views · 🎬 Content Creation & Video
Description
Generate Videos from Images with Wan 2.2 I2V A14B AI Model
🌍 Overview
This workflow connects n8n with Replicate’s Wan-Video model to generate video content from an image + prompt. It handles:
- 🔑 API authentication
- 📤 Sending a generation request
- ⏳ Polling until completion
- 📥 Returning a downloadable video link
🟢 Section 1: Start & Authentication
🔗 Nodes: 1️⃣ On clicking ‘execute’ (Manual Trigger) ▶️ Starts the workflow manually when you click Execute Workflow.
2️⃣ Set API Key 🔑 Stores your Replicate API Key securely inside n8n.
- Value:
YOUR_REPLICATE_API_KEY
💡 Beginner Benefit: ✅ No need to hardcode API keys in every request ✅ Easy to swap keys for different accounts
🔵 Section 2: Send Video Generation Request
🔗 Nodes:
3️⃣ Create Prediction (HTTP Request)
📤 Sends a request to Replicate API → /v1/predictions
-
Model:
wan-video/wan-2.2-i2v-a14b -
Input Parameters:
- 📝
prompt→ your text description - 🖼️
image→ input video/image URL - 🎲
seed→ reproducibility - 🎞️
num_frames→ 81 frames - ⏩
sample_shift→ 5 - ⚙️
sample_steps→ 30 - 🎬
frames_per_second→ 16
- 📝
4️⃣ Extract Prediction ID (Code) 🆔 Pulls out the prediction ID + status from Replicate’s response. This ID is needed for polling until the video finishes generating.
💡 Beginner Benefit: ✅ Automates the API call to generate video ✅ Extracts ID → avoids manual copy-paste
🟣 Section 3: Polling & Status Check
🔗 Nodes: 5️⃣ Wait (2 sec) ⏳ Adds a short delay before checking progress.
- Prevents API spam
6️⃣ Check Prediction Status (HTTP Request)
🔄 Calls Replicate API → /v1/predictions/{id}
- Checks whether the video is still processing or finished
7️⃣ Check If Complete (IF Node)
❓ Compares status → succeeded or not
- ✅ If complete → goes to Process Result
- ❌ If not → loops back to Wait and tries again
💡 Beginner Benefit: ✅ Automatic re-checking → you don’t need to refresh manually ✅ Works until result is ready
🟡 Section 4: Process Final Result
🔗 Nodes: 8️⃣ Process Result (Code) 📥 Once complete, extracts:
- ✅ Status
- 🎞️ Output video URL
- 📊 Generation metrics
- ⏱️ Timestamps (created_at, completed_at)
- 🧩 Model used
Final output = direct video URL you can download/share 🎥
📊 Final Overview
| Section | What Happens | Why It’s Useful |
|---|---|---|
| 🟢 Auth | Manual Trigger + API Key | Secure, easy start |
| 🔵 Send | Create Prediction + Extract ID | Kicks off video generation |
| 🟣 Poll | Wait + Check Status + IF | Auto-checks progress until done |
| 🟡 Result | Process Result | Gives final video link + metadata |
🚀 Why This Workflow Rocks
- 🎥 Video from Image + Prompt → AI-powered video generation in minutes
- 🔄 Fully automated → from request to final video, no manual checks
- 🔑 Secure → API keys handled safely
- ⚙️ Customizable → tweak frames, FPS, prompt, seed
✨ With this workflow, you’ve got a personal AI video generator inside n8n — all you need to do is provide a prompt + input image/video, and you’ll get a fresh AI video link back.
🔗 Nodes Used
HTTP Request
📥 Import
Download workflow.json and import into n8n:
Workflow menu → Import from File