n8n AI Workflows

Production-ready n8n workflows for AI video and image generation. VEO 3, Sora 2, Kling, Seedream, NanoBanana. Built by Copyweb.io

---

What's Inside

A collection of battle-tested n8n workflows powering AI content generation across multiple providers. These workflows handle image generation, video generation, content repurposing, and multi-platform distribution.

Case Studies

System	Description
AI Content Pipeline	Google Sheets to image preview to human approval to video generation
Content Repurposing Engine	Telegram-based pipeline turning existing content into new formats
Motion Ads System	Single product image to 9 motion video variations

Supported Models

Video Generation:

• VEO 3 / VEO 3.1 (Google DeepMind) : talking heads, human movement, keyframe animation
• Sora 2 / Sora 2 Pro (OpenAI) : cinematic motion, smooth camera work
• Kling 2.5 / 3.0 (Kuaishou) : product motion, lip sync
• Wan 2.5 (Alibaba) : fast generation, good for drafts

Image Generation:

• Seedream v4 (ByteDance) : photorealistic product shots
• NanoBanana / NanoBanana Pro : multi-image compositing, enhancement
• Imagen 3 (Google) : high quality, free tier

How to Use

1. Download the .json workflow file
2. In n8n: Workflows > Import from File
3. Select the downloaded file
4. Update credentials (they are not included for security)

Architecture Pattern

All workflows follow the same pattern:

Trigger (webhook/schedule)
  → AI Provider API call
  → Poll for completion
  → Process result
  → Deliver (Airtable / Google Sheets / Telegram / email)

Human review gates are built into every production workflow. Nothing auto-publishes without approval.

---

Built by Copyweb — architecting how marketing teams operate in an AI-first world.