BAMM 2.0
Text to motion
Describe the action in plain language and the model generates the animation.
ResearchMeet the models behind AutoMoGen →
Generative models for human motion.
AutoMoGen is an AI research lab and product company. We build models that turn words, music, and paths into realistic 3D character animation, and the tools to generate, edit, and stream it in production.
Free to start. Sign in with Google, no motion-capture suit required.
Plugins and integrations for the tools you already use →
Blender
Maya
Unity
Unreal Engine
Mixamo
Our models
Direct motion the way you think about it.
Three model families, each built for a different way of describing movement. All of them produce fully retargetable animation.
Read about our research →DanceMosaic
Music to motion
Drop in an 8–12 second audio clip and the character dances to it, on the beat.
MaskControl
Trajectory to motion
Draw a path on the canvas. Motion follows it, with an optional prompt to shape the action.
Products
Generate, edit, and stream motion at scale.
One set of tools from first idea to final take, in the browser, on your desktop, or inside your engine.
Use cases
Built for the way motion actually ships.
From two-person studios to research labs, wherever a motion pipeline gets in the way of the story.
Games
Indie game prototyping
Block out character moves before mocap is on the table, and iterate at the pace of design.
Film & TV
Previsualization
Turn a director's note into shot-ready motion in minutes, not days.
XR
VR and AR characters
Fill immersive worlds with believable motion at scale, without a rigger per character.
Research
Motion research
Use BAMM as a controllable, reproducible baseline for retargeting and synthesis experiments.
Your first animation is one prompt away.
Skip the rigging and keyframes. Type what you want to see and watch it move in seconds.