Craft a consistent character, render it from every angle and scenario, then animate a still into a lip-synced talking-head video - all on one visual node canvas, with swappable AI models.
No account needed to try the free model.
Write the concept and lock the face with a reference image.
Pick every viewpoint and scenario you need in one node.
Generate a consistent set at your chosen ratio and resolution.
Animate a still into a lip-synced talking-head video.
Wire Design → Angles → Render → Motion. Change one node and everything downstream follows.
Pick per project: Gemini, Seedream, FLUX for images; Veo 3.1, Seedance 2.5, Kling, Hailuo for video.
A free, keyless image model lets you explore instantly. Bring your own fal key to unlock the premium models.
Lock a face and carry it across every angle, scenario and environment you generate.
Type a line and the character speaks it with native audio and lip-sync (Veo 3.1 / Seedance 2.5 — one take up to 30s).
Sign in to keep your characters and canvases in your account and pick up on any device.
The actual canvas - nodes wired from concept to render.
Premium generation runs on your own fal.ai account - you're always in control of spend.
Get a fal.ai key →
Type a concept - or drop in a photo - and generate options until one clicks. Then lock the look.

Render your locked character from front, profile, three-quarter and more - identity stays consistent.

Kitchen, café, on stage, outdoors - put your character in any setting in a single click.

Animate a still into video - type a line and your character speaks it with lip-synced audio.
Save your projects and pick up on any device. Generation still uses your own fal key.