Skip to content

LPM 1.0

A 17B model for real-time conversational video

Visit official site

What LPM 1.0 is

The model is described as a 17B-parameter Diffusion Transformer for real-time full-duplex conversational AI video generation.

What you can do with it

  • Condition generation on text, audio and images
  • Keep a character consistent across long clips
  • Work across human, 2D and 3D character styles

Model notes

  • Latency is stated as low as 0.35 second
  • Output is listed at 480P and 720P, 24 fps
  • The material describes zero-shot style generalisation

Pros & cons

✓ What we like

  • Character identity stays consistent over long output
  • 480P and 720P output are listed

! What to watch out for

  • The work is described as an academic research project
  • An annual plan is tied to a priority queue

FAQ

How large is the model?

The material states 17B parameters in a Diffusion Transformer.

What resolutions are named?

480P and 720P at 24 fps are listed for the low latency output.

Last reviewed: 2026-11-15

More AI video generation tools

View all →

How we review