LongCat (Meituan) · video model
LongCat-Video-Avatar 1.6
LongCat-Video-Avatar 1.6 is LongCat (Meituan)'s latest video-generation model in our dataset, released Nov 15, 2026. Open weights (MIT). Meituan LongCat's open-source audio-driven human video model, announced Nov 15, 2026, with MIT-licensed weights on Hugging Face. The team says it improves lip sync, facial and body motion, and long-form consistency over version 1.5, including videos longer than 10 minutes. Verified against vendor sources on Nov 16, 2026.
Reported Last verified · All models · JSON · Markdown
Primary source: https://huggingface.co/meituan-longcat/LongCat-Video-Avatar-1.6. We are the fast, independent index; the vendor’s own page is the source of record for these model facts.
LongCat-Video-Avatar 1.6 at a glance: takes audio, text, image in; returns video.
- ReleasedNov 15, 2026
- LicenceOpen weightsMIT
Key facts
Vendor-reported facts are read from the linked vendor page on the date shown above; we did not run the model. Independently tested means we ran it; no fact on this page is. Editorial is our own classification, not a vendor claim. Where a linked page does not itself state a value, the row says so. How to read provenance.
| Fact | Value | Label | Source |
|---|---|---|---|
| Vendor | LongCat (Meituan) | Editorial | n/a |
| Released | Nov 15, 2026 Date on the official GitHub README ('Nov 15, 2026: We release LongCat-Video-Avatar 1.6'). The Hugging Face model repo was created on Nov 13, 2026 (createdAt 2026-11-13T02:45:33Z). | Vendor-reported | source |
| Status | generally available | Editorial | n/a |
| Input | audio, text, image The model card describes single- and multi-character audio-driven generation, including audio-text-to-video and audio-image-to-video. | Vendor-reported | source |
| Output | video | Vendor-reported | source |
| Licence | MIT (open weights) | Vendor-reported | source |
Where LongCat-Video-Avatar 1.6 sits in our data
- Released Nov 15, 2026 (week 46 of 2026), in November 2026. Other models in our data released that month: Hy4 flash, Qwen4-Max, Gemini 4 Hydrogen, Gemini Decision 1.
- The previous LongCat (Meituan) model we track is LongCat-2.5 (Nov 2, 2026), 13 days earlier.
- All releases by date: AI model releases by month · new AI models this week.
Alternatives to LongCat-Video-Avatar 1.6
Video models from other vendors that share at least one of our groups with LongCat-Video-Avatar 1.6. Comparable on paper, from each model’s sourced facts; we have not run them.
- PAN 2Institute of Foundation Models (MBZUAI) · Both in: Video, Open weights · released Nov 10, 2026
- Pika H3 FlexPika · Both in: Video, Open weights · released Oct 29, 2026
- Gemini Omni 1.2 FlashGoogle · Both in: Video · Input per 1M tokens (text / image / video) $1.5; Output per second of 720p video (approx.) $0.07; Output video per 1M tokens $12 (mixed) · released Nov 5, 2026
- Genie 4Google · Both in: Video · released Oct 27, 2026
How LongCat-Video-Avatar 1.6 differs from LongCat-2.5
Read from the recorded facts of the two models, not from a head-to-head test. We have not run either.
- We have no recorded price for LongCat-Video-Avatar 1.6; LongCat-2.5 has one.
- Input is audio, text, image, against text, image for LongCat-2.5.
- Output is video, against text.
- LongCat-Video-Avatar 1.6 is a video model; LongCat-2.5 is a llm model.
- LongCat-Video-Avatar 1.6 came out 13 days after LongCat-2.5.
- Facts for the other model: LongCat-2.5.
What the sources say
These are claims from the linked pages, not our measurements.
- Vendor claim (Meituan LongCat): more natural lip animation, videos exceeding 10 minutes with less error accumulation, and faster end-to-end inference than LongCat-Video-Avatar 1.5. We did not run the model.
What we tested
Nothing yet. We have not run LongCat-Video-Avatar 1.6 ourselves, so every fact on this page is Reported from the sources below. When we do test it, the scope will be stated here and in our test logs. See methodology.
Not verified, left out
We could not confirm these on a vendor page, so we do not state them.
- The model card reports a human preference Arena of 33,051 votes in which LongCat-Video-Avatar 1.6 ranks second of six systems with a Bradley-Terry Elo of 1068.1, behind OmniHuman-2.0. These are the team's own evaluation figures. We did not verify them.
- A parameter count for the 1.6 avatar checkpoint is not stated on the Hugging Face model card or in the Nov 15 GitHub announcement. We did not copy the 13.6B figure from the earlier LongCat-Video foundation model.
- The project page's ethical-considerations note says some demo images and audio come from real videos and that those generated demo materials are for academic use. The model card itself states an MIT license for the weights.
LongCat-Video-Avatar 1.6 FAQ
When was LongCat-Video-Avatar 1.6 released?
LongCat (Meituan) released LongCat-Video-Avatar 1.6 on Nov 15, 2026. Source.
Is LongCat-Video-Avatar 1.6 open weights?
Yes. The weights are published under MIT. Source.
Has Agentic Weekly tested LongCat-Video-Avatar 1.6?
Not yet. Every fact on this page is Reported: read from the linked vendor pages, not measured by us.
Sources
Compare and guides
We have no head-to-head page or guide that names LongCat-Video-Avatar 1.6 yet. These hubs cover the models we do compare.
All model comparisons · All guides · Model releases by month
More from LongCat (Meituan)
- LongCat-2.5 (Nov 2, 2026)
Newsletter
The Weekly Verdict. Free, every Monday.
A free email every Monday: the week's few big AI releases, then 5 short use cases for the new models, each with a prompt you can paste.
Free. Double opt-in. One click to unsubscribe.