Pinned Tweet
Introducing Character World Model-1
the most controllable interactive avatar model in the world
Features and how we built it 👇️
Introducing the LemonSlice Cartoon Model.
- Designed to bring cartoon characters to life with maximum animation
- Supports everything from 3D mascots to anime
- Highly expressive interactive avatars from a single image
Available today on all subscription plans!
Here's Muse vs the new CWM-1 model we released yesterday: lemonslice.com/cwm-1
LemonSlice retweeted
Motion design for emotion engine illustration on our blog
The demo video shows our new Emotion Engine. It's from a single take. And, it's fantastic!
The Emotion Engine models the internal state of the avatar, and then predicts what the avatar should be doing (nodding, fidgeting, explaining, flirting, etc).
Part of solving the Avatar Turing test is giving avatars an internal, emotional state. That's why we built this.
LemonSlice retweeted
Replying to @sidney_primas
@sidney_primas @lina_colucci @LemonSliceAI did you all not get the memo to pace the frontier?!
Congrats on the impressive leap! 🤯
LemonSlice retweeted
Replying to @LemonSliceAI
This isnt some insanely expensive pre-recorded demo, try it on our launch page now lemonslice.com/cwm-1
LemonSlice retweeted
Had fun animating a bouncy bar that helps you center your face for our demo (lemonslice.com/cwm-1)
LemonSlice retweeted
We just shipped the first world model you can talk to. Unlike other avatar models, CWM-1 provides full controllability over an avatar's emotions and actions - all in real-time. Hats off to the team!
5/ CWM-1 supports whole-body actions and hand-object interactions on any character with ZERO training.
6/ To build CWM-1, we had to solve multiple hard problems, from training our own video DiT to solving error accumulation for multi-hour long streams.
We detail these and more in our research page: lemonslice.com/cwm-1
5/ CWM-1 supports whole-body actions and hand-object interactions on any character with ZERO training.
6/ To build CWM-1, we had to solve multiple hard problems, from training our own video DiT to solving error accumulation for multi-hour long streams.
We detail these and more in our research page: lemonslice.com/cwm-1










