Auto-generate and publish AI short videos with Sora 2, Veo 3.1, Gemini & Blotato

Enter a text idea in n8n → Gemini rewrites it into a video prompt → parallel calls to Sora 2 Pro and Veo 3.1 → auto-publish to multiple platforms via Blotato

Who

Content creators, brand operators, and UGC teams who need high-frequency AI video production and cross-platform publishing

You get

One chat message triggers video generation from two models, auto-distribution to YouTube, TikTok, and Instagram, plus email notifications

You need

n8n · gemini · gmail · google-sheets

Steps

1. Input idea and enhance prompt

   Send a text idea through the Chat Trigger. n8n passes the input to the Gemini Prompt Enhancer (AI Agent node, linked to Google Gemini Chat Model). Its systemMessage expands the user input into a detailed video prompt suitable for Sora 2 and Veo 3.1, covering action, scene, lighting, mood, camera movement, and style. Output is plain text, shared by both downstream branches.

   Execution prompt · Source text。Gemini Prompt Enhancer node systemMessage field

   Copy prompt

   You are an AI assistant that rewrites user ideas into highly detailed AI video generation prompts.
   
   Rules:
   - Take the user input and expand it into a single improved prompt suitable for models like OpenAI Sora 2 and Google Veo 3.1.
   - Focus on describing action, scene, lighting, mood, camera movement and style.
   - Make it safe for all audiences.
   - Output ONLY the improved prompt as plain text (no JSON, no keys).

2. Configure toggles and credentials

   In the Config – Toggles node (Set node), set booleans: useSora, useVeo, publishToYouTube, publishToTikTok, publishToInstagram, logToGoogleSheets. Then add credentials in n8n: googlePalmApi (Gemini), httpHeaderAuth (Sora and Wavespeed), blotatoApi (Blotato). Replace the placeholder accountId in Blotato nodes with your actual account ID. Fill documentId and sheetName in the Google Sheets node.

3. Sora branch: generate, download, and publish to multiple platforms

   If useSora is true, the flow enters the Sora branch. The Create Sora 2 Pro Video node POSTs to https://api.openai.com/v1/videos with parameters: model=sora-2-pro, size=1280x720, seconds=8, prompt=enhanced prompt. The returned id is used by Download Sora Video to fetch the MP4. Upload Sora Video to Blotato uploads the media, then Publish Sora to YouTube/TikTok/Instagram publishes via Blotato (postContentText and postContentMediaUrls from upstream). Email Sora Upload Confirmation sends a confirmation email, and Log Sora Video to Google Sheets optionally records metadata.

4. Veo branch: generate, wait, retrieve result, and notify

   If useVeo is true, the flow enters the Veo branch. The Create Veo 3.1 Video node POSTs to https://api.wavespeed.ai/api/v3/google/veo3.1/text-to-video with parameters: aspect_ratio=9:16, duration=8, generate_audio=true, resolution=720p, prompt=enhanced prompt. Wait for Veo Result pauses 60 seconds, then Get Veo Result GETs /predictions/{data.id}/result for the output link. Email Veo Video Link emails the link. Optional: Upload Veo Video to Blotato → Publish Veo to YouTube for upload and publishing.

5. Optional extensions and customization

   Suggested extensions from the source: add a title/description generator for YouTube Shorts; insert a thumbnail generator (image AI model); extend logging with Sheets or a database; add additional platforms supported by Blotato; use different prompt strategies for cinematic, viral, or niche content styles. All are optional and do not affect the core workflow.

You are reproducing n8n template 11276: Sora 2 + Veo 3.1 + Gemini + Blotato automated video publishing. First confirm you have the following credentials: Google Gemini (googlePalmApi), OpenAI Sora 2 (httpHeaderAuth), Wavespeed Veo 3.1 (httpHeaderAuth), Blotato (blotatoApi), Gmail OAuth2. Check that the Config – Toggles match your publishing needs, and replace the placeholder accountId in Blotato nodes with your real account ID. Test the Chat Trigger with a simple prompt first to verify the Gemini Prompt Enhancer output format, then enable the Sora/Veo branches.