LongCat Video plans include extra room for longer continuation tests and multi-pass review runs.
10:00:00cs

LongCat AI: The Open-Source Model Lineup

What LongCat AI is, and how its models fit together: LongCat Video (13.6B), LongCat Image, LongCat Avatar, and the separate LongCat-Flash language model.

"LongCat AI" is one name for several different things, and mixing them up is the quickest way to download the wrong weights or follow the wrong tutorial. LongCat is an open-source AI generation framework developed by Meituan. Under that banner sit at least three generation models — a video model, an image model, and an audio-driven avatar model — plus a separately named language model that has nothing to do with video generation. This page maps the lineup so you can pick the right model before you spend an afternoon on the wrong one.

The model lineup at a glance

ModelWhat it generatesWhat it is for
LongCat VideoVideo, up to 961 framesText-to-video, image-to-video, and video continuation
LongCat ImageStill imagesImage generation that runs on modest VRAM
LongCat Video Avatar 1.5Talking-head videoAudio-driven lip-synced avatars from a portrait + voice clip
LongCat-FlashTextA separate 560B-parameter language model for reasoning, not video

The critical separation: LongCat-Flash is a language model, not a video model. If you want reasoning over text, that is the name to reach for. If you want video or images, you are dealing with the generation models above.

What LongCat Video brings to the table

LongCat Video was released by Meituan in September 2025 under an MIT license. It is a 13.6 billion parameter Dense Transformer built around one idea: temporal coherence over long sequences, where most short-clip models start to drift:

  • Up to 961 frames of output, far beyond the 5–10 second ceiling of short-clip tools
  • Text-to-video and image-to-video in a single model
  • Video continuation — extend an existing sequence while carrying subject identity and motion forward
  • 480p or 720p output, sized for real edits rather than previews

The 13.6B figure and the frame ceiling come straight from the model's published description; the differentiator worth remembering is continuation, which is what makes minutes-long shots practical instead of a slideshow of glued clips.

How LongCat Image differs

LongCat Image is the still-image sibling. The relevant practical point is the opposite of LongCat Video's: where the video model is heavy and long-form, the image model is positioned to run on lower-VRAM hardware. It is not the same checkpoint as the video model — download the image weights for image work, and do not assume one download covers both.

When you want an avatar instead

LongCat Video Avatar 1.5 is the audio-driven path. Feed it a portrait and a short WAV, and it generates a lip-synced talking head. Its audio encoder and inference path differ from the base video model, so treat it as its own tool rather than a setting on LongCat Video. This is the model behind faceless presenters and recurring virtual hosts.

Picking the right model: a decision list

  1. Text-to-video or image-to-video? → LongCat Video
  2. Need to extend an existing clip without reseeding the look? → LongCat Video continuation
  3. Need a still image on a modest GPU? → LongCat Image
  4. Need a face that speaks a voice clip? → LongCat Video Avatar 1.5
  5. Need text reasoning or a long-context language model? → LongCat-Flash, not any of the generation models
  6. Need minutes of coherent footage? → LongCat Video, because continuation is the only path that holds identity over time
  7. Open-source with an MIT license for the video model? → yes for LongCat Video, but always re-check the license of each specific checkpoint and your inputs before commercial use

The open-source caveat

Open weights are not a commercial license. "MIT" and "open source" describe how you may use and modify the code and weights; they do not automatically grant rights to the portrait in your avatar, the music in your video, or a third-party checkpoint you layered on top. Review the license of each model you download and the rights attached to every input before you publish or sell the result.

Explore the lineup from the LongCat Video homepage.

Frequently Asked Questions

What is LongCat AI?

LongCat is an open-source AI generation framework developed by Meituan. It includes a video model (LongCat Video), an image model (LongCat Image), and an audio-driven avatar model (LongCat Video Avatar 1.5).

Is LongCat-Flash a video model?

No. LongCat-Flash is a separate 560-billion-parameter language model for text reasoning. It has nothing to do with LongCat Video or image generation, despite sharing the LongCat name.

How many parameters does LongCat Video have?

LongCat Video is a 13.6 billion parameter Dense Transformer. It generates up to 961 frames at 480p or 720p and supports text-to-video, image-to-video, and video continuation.

What is the difference between LongCat Image and LongCat Video?

LongCat Video generates motion — clips and continued sequences. LongCat Image generates still images and is positioned to run on lower-VRAM hardware. They are separate checkpoints; one download does not cover both.

Does "open source" mean I can use the output commercially?

Not by itself. Open weights describe how you may use the model, not the rights to your inputs or to third-party checkpoints. Review each model license and the rights to every portrait, audio, and dependency before commercial use.

Start From the Right LongCat Model

Video, image, avatar, or reasoning — pick the correct checkpoint and skip the wrong-tutorial rabbit hole.