ZMIME
Agnes AI · video model · Released Nov 20, 2026

Agnes Video H3 MoE

Experimental MiniMax H3 FL2VA adaptation that replaces selected feed-forward layers with routed sparse experts.

  • Video model
  • Open weights
  • Text to video
  • Image to video
  • Native audio
  • Preview
Arena score
—Not enough votes yet
Starting price
—Not published
Longest clip
—Not published
Resolution
—With native audio
Quality & price

How it rates and what it costs

Agnes Video H3 MoE does not have enough arena votes for a quality score yet.

Specifications

The details

Lab
Agnes AI
Task
Audiovisual generation
Published capabilities
Experimental MiniMax H3 FL2VA adaptation that replaces selected feed-forward layers with routed sparse experts.
Access
Public composite package with original weights, learned router and adapter, and reference loader
Result
Video with synchronized audio
Operating limits
Supports an initial frame, an ending frame or both with the publisher’s reference loader.
Type
Video model
Released
Nov 20, 2026
Inputs
Text prompts, Reference images
Output
Video with synchronized audio
Can do
Text to video, Image to video, Native audio
Sound
Always generated
Frame control
First and last frame
Availability
Not on the API hosts we track
Status
Preview

Agnes AI documentation

Questions

About Agnes Video H3 MoE

Who makes Agnes Video H3 MoE?

Agnes Video H3 MoE is a video model from Agnes AI, released Nov 20, 2026.

What can Agnes Video H3 MoE do?

Agnes Video H3 MoE accepts text prompts and reference images and produces video with synchronized audio. In practice: text to video, image to video and native audio.

Is Agnes Video H3 MoE open source?

Yes. Agnes AI publishes the weights under the MiniMax H3 Community License, so you can run it on your own hardware.

Where can I use Agnes Video H3 MoE?

Public composite package with original weights, learned router and adapter, and reference loader

What are the alternatives to Agnes Video H3 MoE?

The closest video models from other labs, by release date, are H3-Tiny-17B-Pruned-FL2VA (H3-Tiny, Nov 20, 2026), AURA 1.3B (Kyutai, Nov 20, 2026), LongCat-Video-Avatar 1.6 (Meituan (LongCat), Nov 15, 2026) and PAN 2 (MBZUAI Institute of Foundation Models, Nov 10, 2026).