Yaron Been This workflow corresponds to n8n.io template #7012 — we link there as the canonical source.
The workflow JSON
Copy or download the full n8n JSON below. Paste it into a new n8n workflow, add your credentials, activate. Full import guide →
{
"id": "yWCiA40cYIUrujSC",
"name": "14. Generate Images from Text with IBM Granite Vision 3.3 2B AI Model",
"tags": [],
"nodes": [
{
"id": "4329b4aa-8904-456b-a382-cdb1eb6831b8",
"name": "On clicking 'execute'",
"type": "n8n-nodes-base.manualTrigger",
"position": [
64,
144
],
"parameters": {},
"typeVersion": 1
},
{
"id": "72600d86-2ff3-46df-ae23-335cb01f0e59",
"name": "Set API Key",
"type": "n8n-nodes-base.set",
"position": [
288,
144
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "64cd57c7-2224-48a6-88df-c37b310f8eb3",
"name": "replicate_api_key",
"type": "string",
"value": "YOUR_REPLICATE_API_KEY"
}
]
}
},
"typeVersion": 3.3
},
{
"id": "a1310c30-6206-4016-a0c3-e32e6a0524ba",
"name": "Create Prediction",
"type": "n8n-nodes-base.httpRequest",
"position": [
512,
144
],
"parameters": {
"url": "https://api.replicate.com/v1/predictions",
"method": "POST",
"options": {
"timeout": 60000
},
"jsonBody": "{\n \"version\": \"3339e8453ca94104383f6f085a511d7f26cca2d0cab2f6018986737b6cf7d391\",\n \"input\": {\n \"seed\": 1,\n \"top_k\": 50,\n \"top_p\": 0.9,\n \"max_tokens\": 512,\n \"temperature\": 0.6\n }\n}",
"sendBody": true,
"sendHeaders": true,
"specifyBody": "json",
"authentication": "genericCredentialType",
"genericAuthType": "httpHeaderAuth",
"headerParameters": {
"parameters": [
{
"name": "Authorization",
"value": "={{ 'Bearer ' + $('Set API Key').item.json.replicate_api_key }}"
},
{
"name": "Content-Type",
"value": "application/json"
}
]
}
},
"typeVersion": 4.2
},
{
"id": "c79ea010-0525-4c61-8ca7-bdbcc84b426f",
"name": "Extract Prediction ID",
"type": "n8n-nodes-base.code",
"position": [
720,
144
],
"parameters": {
"mode": "runOnceForEachItem",
"jsCode": "// Extract prediction ID and initial status\\nconst prediction = $input.item.json;\\nconst predictionId = prediction.id;\\nconst initialStatus = prediction.status;\\n\\n// Store the prediction ID for polling\\nreturn {\\n predictionId: predictionId,\\n status: initialStatus,\\n predictionUrl: `https://api.replicate.com/v1/predictions/${predictionId}`\\n};"
},
"typeVersion": 2
},
{
"id": "d8fc6cb6-f37f-4c45-b005-83407697ca34",
"name": "Wait",
"type": "n8n-nodes-base.wait",
"position": [
944,
144
],
"parameters": {
"unit": "seconds",
"amount": 2
},
"typeVersion": 1
},
{
"id": "53f39db2-e688-46ce-9a3d-b50c2365d11b",
"name": "Check Prediction Status",
"type": "n8n-nodes-base.httpRequest",
"position": [
1168,
144
],
"parameters": {
"url": "={{ $json.predictionUrl }}",
"options": {},
"sendHeaders": true,
"authentication": "genericCredentialType",
"genericAuthType": "httpHeaderAuth",
"headerParameters": {
"parameters": [
{
"name": "Authorization",
"value": "={{ 'Bearer ' + $('Set API Key').item.json.replicate_api_key }}"
}
]
}
},
"typeVersion": 4.2
},
{
"id": "32f0b031-596d-4e32-bc58-a002a46a2c29",
"name": "Check If Complete",
"type": "n8n-nodes-base.if",
"position": [
1392,
144
],
"parameters": {
"conditions": {
"boolean": [
{
"value1": "={{ $json.status }}",
"value2": "succeeded"
}
]
}
},
"typeVersion": 1
},
{
"id": "f9e3961d-ee2f-47e7-9b61-23f694091cca",
"name": "Process Result",
"type": "n8n-nodes-base.code",
"position": [
1600,
48
],
"parameters": {
"mode": "runOnceForEachItem",
"jsCode": "// Process the completed prediction\\nconst result = $input.item.json;\\n\\nreturn {\\n status: result.status,\\n output: result.output,\\n metrics: result.metrics,\\n created_at: result.created_at,\\n completed_at: result.completed_at,\\n model: 'ibm-granite/granite-vision-3.3-2b',\\n image_url: result.output\\n};"
},
"typeVersion": 2
},
{
"id": "776eb5ba-cfc1-4505-9ba9-35fa7b10a6d2",
"name": "Sticky Note",
"type": "n8n-nodes-base.stickyNote",
"position": [
16,
-288
],
"parameters": {
"color": 3,
"width": 406,
"height": 639,
"content": "## \ud83d\udfe2 Section 1: Trigger & API Setup\n\n\ud83d\udd17 **Nodes:**\n\n* **Manual Trigger** \u2192 Starts when you click *Execute*.\n* **Set API Key** \u2192 Stores your Replicate API Key safely in the workflow.\n\n\ud83d\udca1 **Beginner takeaway:**\nThis section is like turning the key in the ignition. You start the workflow, and it loads your credentials so you can talk to Replicate\u2019s API.\n\n\ud83d\udcc8 **Advantage:**\nKeeps your API key stored inside the workflow instead of hard-coding it everywhere.\n\n---\n\n"
},
"typeVersion": 1
},
{
"id": "8f6c9ce5-1c0c-4d66-94e3-d4d8e2dfdbb5",
"name": "Sticky Note9",
"type": "n8n-nodes-base.stickyNote",
"position": [
-1760,
-480
],
"parameters": {
"color": 4,
"width": 1300,
"height": 320,
"content": "=======================================\n WORKFLOW ASSISTANCE\n=======================================\nFor any questions or support, please contact:\n Yaron@nofluff.online\n\nExplore more tips and tutorials here:\n - YouTube: https://www.youtube.com/@YaronBeen/videos\n - LinkedIn: https://www.linkedin.com/in/yaronbeen/\n=======================================\n"
},
"typeVersion": 1
},
{
"id": "09a0a691-f7ac-4d4d-b90e-d33e5c8d3f92",
"name": "Sticky Note4",
"type": "n8n-nodes-base.stickyNote",
"position": [
-1760,
-144
],
"parameters": {
"color": 4,
"width": 1289,
"height": 2286,
"content": "# Generate Images from Text with IBM Granite Vision 3.3 2B AI Model\n\n### \ud83c\udf0d **Overview**\n\nThis workflow uses the **ibm-granite/granite-vision-3.3-2b** model (hosted on Replicate) to generate AI images. It starts manually, sends a request to the Replicate API, waits for the result, and finally outputs the generated image link.\n\nThink of it as your **AI art assistant** \u2014 you click once, and it handles the full request/response cycle for image generation.\n\n---\n\n## \ud83d\udfe2 Section 1: Trigger & API Setup\n\n\ud83d\udd17 **Nodes:**\n\n* **Manual Trigger** \u2192 Starts when you click *Execute*.\n* **Set API Key** \u2192 Stores your Replicate API Key safely in the workflow.\n\n\ud83d\udca1 **Beginner takeaway:**\nThis section is like turning the key in the ignition. You start the workflow, and it loads your credentials so you can talk to Replicate\u2019s API.\n\n\ud83d\udcc8 **Advantage:**\nKeeps your API key stored inside the workflow instead of hard-coding it everywhere.\n\n---\n\n## \ud83d\udfe6 Section 2: Create Prediction\n\n\ud83d\udd17 **Nodes:**\n\n* **HTTP Request (Create Prediction)** \u2192 Sends a request to Replicate with the chosen model (`granite-vision-3.3-2b`) and input parameters (seed, temperature, max\\_tokens, etc.).\n\n\ud83d\udca1 **Beginner takeaway:**\nThis is where the workflow actually **asks the AI model to generate an image**.\n\n\ud83d\udcc8 **Advantage:**\nYou can tweak parameters like **creativity (temperature)** or **randomness (seed)** to control results.\n\n---\n\n## \ud83d\udfe3 Section 3: Polling & Status Check\n\n\ud83d\udd17 **Nodes:**\n\n* **Extract Prediction ID (Code)** \u2192 Saves the unique job ID.\n* **Wait (2s)** \u2192 Pauses before checking status.\n* **Check Prediction Status (HTTP Request)** \u2192 Calls Replicate to see if the image is ready.\n* **If Condition (Check If Complete)** \u2192\n\n * \u2705 If `status = succeeded` \u2192 move to result\n * \ud83d\udd04 Else \u2192 go back to Wait and check again\n\n\ud83d\udca1 **Beginner takeaway:**\nSince image generation takes a few seconds, this section keeps **asking the AI \u201care you done yet?\u201d** until the image is ready.\n\n\ud83d\udcc8 **Advantage:**\nNo need to guess \u2014 the workflow waits automatically and retries until success.\n\n---\n\n## \ud83d\udd35 Section 4: Process Result\n\n\ud83d\udd17 **Nodes:**\n\n* **Process Result (Code)** \u2192 Extracts the final data:\n\n * \u2705 Status\n * \u2705 Output image URL\n * \u2705 Metrics (time taken, etc.)\n * \u2705 Model info\n\n\ud83d\udca1 **Beginner takeaway:**\nThis section **collects the finished image link** and prepares it neatly for you.\n\n\ud83d\udcc8 **Advantage:**\nYou get structured output that you can **save, display, or use in another workflow** (like auto-sending images to Slack or saving to Google Drive).\n\n---\n\n## \ud83d\udcca Final Overview Table\n\n| Section | Nodes | Purpose | Benefit |\n| -------------------- | ---------------------------------- | --------------------------- | --------------------------- |\n| \ud83d\udfe2 Trigger & Setup | Manual Trigger, Set API Key | Start + load credentials | Secure API key management |\n| \ud83d\udfe6 Create Prediction | HTTP Request | Ask AI to generate image | Control creativity & output |\n| \ud83d\udfe3 Polling | Extract ID, Wait, Check Status, If | Repeatedly check job status | Auto-wait until done |\n| \ud83d\udd35 Process Result | Process Result | Extract image + details | Get clean output for reuse |\n\n---\n\n## \ud83d\ude80 Why This Workflow is Useful\n\n* **Automates full API cycle** \u2192 From request to final image URL\n* **Handles delays automatically** \u2192 Keeps checking until your image is ready\n* **Customizable parameters** \u2192 Adjust creativity, randomness, and token limits\n* **Reusable** \u2192 Connect it to email, Slack, Notion, or storage for instant sharing\n* **Beginner-friendly** \u2192 Just plug in your API key and hit *Execute*\n\n---\n"
},
"typeVersion": 1
},
{
"id": "6c367d94-4ec5-44f8-9a90-4e3e2df00608",
"name": "Sticky Note1",
"type": "n8n-nodes-base.stickyNote",
"position": [
448,
-464
],
"parameters": {
"color": 4,
"width": 230,
"height": 815,
"content": "## \ud83d\udfe6 Section 2: Create Prediction\n\n\ud83d\udd17 **Nodes:**\n\n* **HTTP Request (Create Prediction)** \u2192 Sends a request to Replicate with the chosen model (`granite-vision-3.3-2b`) and input parameters (seed, temperature, max\\_tokens, etc.).\n\n\ud83d\udca1 **Beginner takeaway:**\nThis is where the workflow actually **asks the AI model to generate an image**.\n\n\ud83d\udcc8 **Advantage:**\nYou can tweak parameters like **creativity (temperature)** or **randomness (seed)** to control results.\n\n---\n\n"
},
"typeVersion": 1
},
{
"id": "7b765db4-0283-47d4-8ab9-df75aa64f6ad",
"name": "Sticky Note2",
"type": "n8n-nodes-base.stickyNote",
"position": [
704,
-336
],
"parameters": {
"color": 6,
"width": 806,
"height": 687,
"content": "## \ud83d\udfe3 Section 3: Polling & Status Check\n\n\ud83d\udd17 **Nodes:**\n\n* **Extract Prediction ID (Code)** \u2192 Saves the unique job ID.\n* **Wait (2s)** \u2192 Pauses before checking status.\n* **Check Prediction Status (HTTP Request)** \u2192 Calls Replicate to see if the image is ready.\n* **If Condition (Check If Complete)** \u2192\n\n * \u2705 If `status = succeeded` \u2192 move to result\n * \ud83d\udd04 Else \u2192 go back to Wait and check again\n\n\ud83d\udca1 **Beginner takeaway:**\nSince image generation takes a few seconds, this section keeps **asking the AI \u201care you done yet?\u201d** until the image is ready.\n\n\ud83d\udcc8 **Advantage:**\nNo need to guess \u2014 the workflow waits automatically and retries until success.\n\n---\n\n"
},
"typeVersion": 1
},
{
"id": "8fbd6a2c-dcfb-4723-8d4e-caf5c3a563ca",
"name": "Sticky Note3",
"type": "n8n-nodes-base.stickyNote",
"position": [
1536,
-576
],
"parameters": {
"color": 5,
"width": 230,
"height": 927,
"content": "## \ud83d\udd35 Section 4: Process Result\n\n\ud83d\udd17 **Nodes:**\n\n* **Process Result (Code)** \u2192 Extracts the final data:\n\n * \u2705 Status\n * \u2705 Output image URL\n * \u2705 Metrics (time taken, etc.)\n * \u2705 Model info\n\n\ud83d\udca1 **Beginner takeaway:**\nThis section **collects the finished image link** and prepares it neatly for you.\n\n\ud83d\udcc8 **Advantage:**\nYou get structured output that you can **save, display, or use in another workflow** (like auto-sending images to Slack or saving to Google Drive).\n\n---\n\n"
},
"typeVersion": 1
}
],
"active": false,
"settings": {
"executionOrder": "v1"
},
"versionId": "0fac2de0-55b2-4dcb-bf9b-d36d42f23ea2",
"connections": {
"Wait": {
"main": [
[
{
"node": "Check Prediction Status",
"type": "main",
"index": 0
}
]
]
},
"Set API Key": {
"main": [
[
{
"node": "Create Prediction",
"type": "main",
"index": 0
}
]
]
},
"Check If Complete": {
"main": [
[
{
"node": "Process Result",
"type": "main",
"index": 0
}
],
[
{
"node": "Wait",
"type": "main",
"index": 0
}
]
]
},
"Create Prediction": {
"main": [
[
{
"node": "Extract Prediction ID",
"type": "main",
"index": 0
}
]
]
},
"Extract Prediction ID": {
"main": [
[
{
"node": "Wait",
"type": "main",
"index": 0
}
]
]
},
"On clicking 'execute'": {
"main": [
[
{
"node": "Set API Key",
"type": "main",
"index": 0
}
]
]
},
"Check Prediction Status": {
"main": [
[
{
"node": "Check If Complete",
"type": "main",
"index": 0
}
]
]
}
}
}
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
This workflow uses the ibm-granite/granite-vision-3.3-2b model (hosted on Replicate) to generate AI images. It starts manually, sends a request to the Replicate API, waits for the result, and finally outputs the generated image link.
Source: https://n8n.io/workflows/7012/ — original creator credit. Request a take-down →
Related workflows
Workflows that share integrations, category, or trigger type with this one. All free to copy and import.
This workflow allows you to import any workflow from a file or another n8n instance and map the credentials easily. A multi-form setup guides you through the entire process At the beginning you have t
Execute Command, Read Write File, HTTP Request +3
[n8n] Advanced URL Parsing and Shortening Workflow - Switchy.io Integration. Uses splitInBatches, stickyNote, httpRequest, html. Event-driven trigger; 56 nodes.
HTTP Request, GitHub, Stop And Error +1
[](https://youtu.be/c7yCZhmMjtI)
HTTP Request, GitHub, Stop And Error +1
This automation organizes your n8n workflows files into categorizes (Active, Template, Done, Archived) and uploads them directly to a categorized Google Drive folders. It is designed to help users man
Google Drive, HTTP Request, Time Saved
Create Animated Stories using GPT-4o-mini, Midjourney, Kling and Creatomate API. Uses httpRequest. Event-driven trigger; 51 nodes.
HTTP Request