Ingest a research paper, run a chain of image edits, then merge, branch and exclude concepts on one visual canvas. Explore real creations below — sign in when you're ready to make your own.
No credit card to explore · Sign in with Google or Discord to create.
Describe an object or drop a photo — Grafy builds a textured 3D mesh you can rig, animate, remesh, print-prep, view in AR and export.
Free to start · sign in to generate.
Handy browser tools that run entirely on your device, so nothing is uploaded. Then sign in to generate your own 3D models.
Powered by frontier models from the world's leading AI labs
A look at what creators produce inside Grafy — every card is a saved session you can branch from once you're signed in.
Drop in a paper and Grafy extracts the key concepts and how they relate, so you can explore the architecture instead of re-reading 15 pages.
Stack reversible edit steps and keep every version on the canvas — branch off any step without losing the original.
Combine two research threads into one map, or prune nodes you don't need — every operation is non-destructive and reversible.
Draft, summarize and reason with open models — grounded in the documents you ingest.
Generate from a prompt or transform an existing image through reversible edit steps.
Turn a still or a concept into a short clip and keep it on the same canvas.
Add a PDF, an image or a prompt to a fresh canvas.
Run edits, generations and extractions as reversible steps.
Merge, branch and exclude nodes to shape the graph.
Share the session or export the result when it's ready.
Grafy is a multimodal creation studio that turns documents, images and ideas into editable knowledge graphs. Ingest a PDF, generate or edit images and video, then merge, branch and exclude concepts on one visual canvas.
Upload a PDF and Grafy extracts its key concepts and how they relate, then lays them out as a navigable graph you can explore, branch and question - instead of re-reading the whole document.
Browsing the public gallery is free, and you can create a free account to start building. Paid plans add more generation credits and higher limits for image, video and audio workflows.
Images and multi-step image edits, short videos, audio and music, grounded text, and document knowledge graphs - all on one canvas where every step is a node you can revisit.
Grafy routes each task to the best-suited model from the world’s leading AI labs, and can also run local open models — so you’re never locked to a single vendor.
Instead of a linear chat, Grafy is a visual graph: every generation, edit and extraction becomes a node you can branch, merge, compare and exclude, so complex multi-step work stays organized and reversible.
Yes — Grafy occasionally shares promo codes that add free credits. Redeem them under My account, in the “Redeem a promo code” box, and the credits are added to your balance instantly.
Browsing the gallery is free. To run a workflow on your own files, create a free account — it takes a few seconds with Google or Discord.