ZMIME
Kyutai · video model · Released Nov 20, 2026

AURA 1.3B

A four-step causal model animates a portrait from streaming speech, rendering chunks from the available audio.

  • Video model
  • Open weights
  • Image to video
  • Audio input
Arena score
—Not enough votes yet
Starting price
—Not published
Longest clip
—Not published
Resolution
576pAudio unconfirmed
Quality & price

How it rates and what it costs

AURA 1.3B does not have enough arena votes for a quality score yet.

Specifications

The details

Lab
Kyutai
Task
Speech-driven portrait animation
Published capabilities
A four-step causal model animates a portrait from streaming speech, rendering chunks from the available audio.
Model size
1.3B parameters
Access
Public student checkpoints and inference code
Operating limits
Kyutai reports 20.7 fps end to end on one H100 at 576p for 16 Hz playback.
Type
Video model
Released
Nov 20, 2026
Inputs
Reference images, Audio
Output
Video
Can do
Image to video, Audio input
Resolutions
576p
Availability
Not on the API hosts we track

Kyutai documentation

Same lab

More video models from Kyutai

Questions

About AURA 1.3B

Who makes AURA 1.3B?

AURA 1.3B is a video model from Kyutai, released Nov 20, 2026.

What can AURA 1.3B do?

AURA 1.3B accepts reference images and audio and produces video. In practice: image to video and audio input.

Is AURA 1.3B open source?

Yes. Kyutai publishes the weights under the CC BY-NC 4.0, so you can run it on your own hardware.

Where can I use AURA 1.3B?

Public student checkpoints and inference code

What are the alternatives to AURA 1.3B?

The closest video models from other labs, by release date, are AURA 5B (Kyutai, Nov 20, 2026), H3-Tiny-17B-Pruned-FL2VA (H3-Tiny, Nov 20, 2026), LongCat-Video-Avatar 1.6 (Meituan (LongCat), Nov 15, 2026), PAN 2 (MBZUAI Institute of Foundation Models, Nov 10, 2026) and Viggle Remake (Viggle AI, Nov 2, 2026).