A node-based canvas to chain LLMs, image and audio generation, and video generation — plus rich inputs, an output hub, and selective media export. Connect to hundreds of models through OpenRouter. No code required.
Connect models, chain outputs, and generate content — all from a visual canvas.
Connect to any text model on OpenRouter. GPT-4o, Claude, Gemini, Llama, and hundreds more.
Generate images with DALL-E, Flux, Stable Diffusion. Text-to-image and image-to-image.
Create videos with Veo 3.1, Sora 2, Seedance. Full control over duration, resolution, and audio.
Generate speech and other audio from models with audio output on OpenRouter — wire prompts or LLM text into an Audio Gen node.
Upload files or paste URLs for images and videos. Feed vision models, references, or frame-to-video workflows.
Download canvas media as a ZIP with a dialog to pick what to include: inputs, image/video generators, and output previews.
Drag, drop, and connect nodes to build complex AI workflows. No code required.
Execute your entire pipeline with a single button. Nodes run in dependency order automatically.
Save workflows as JSON files. Share pipelines with your team or the community.
Self-host for free or try the cloud beta. You always bring your own OpenRouter API key.
Run on your own machine. Full control, no limits.
Hosted Studio in private beta — sign in and build without installing anything.
Start building AI pipelines in minutes. Free to self-host, or sign up for cloud access.