AgentScout / video generation / HunyuanVideo
video generationOpen weightsNewAdded Oct 18, 2026

HunyuanVideo logoHunyuanVideo

No ratings yet★ Rate this tool

Tencent's open large video generation model and framework.

Updated

✨ Try AgentScout's free AI image generator →

Use cases

About HunyuanVideo

HunyuanVideo is listed in our video generation category and is aimed at builders. Tencent's open large video generation model and framework. Pricing, as the maker states it: Open weights (Tencent Hunyuan Community License; not valid in the EU, UK and South Korea). Maturity: Active. We last checked the official link on 2026-10-20.

What is HunyuanVideo?

HunyuanVideo is Tencent's open video foundation model and framework for text-to-video generation, released with PyTorch model code, pre-trained weights and inference code. At release Tencent described it as the largest open video generation model, with over 13 billion parameters, and reported quality comparable to leading closed models.

It is integrated into Hugging Face Diffusers, has FP8 weights to save GPU memory and parallel inference via xDiT. Tencent has since built a family on it: HunyuanVideo-I2V (image-to-video), HunyuanCustom, HunyuanVideo-Avatar (audio-driven animation) and the newer, more efficient HunyuanVideo-1.5.

Sources: the official README and our directory listing, checked 2026-10-20. Official site: github.com/Tencent-Hunyuan/HunyuanVideo.

Who is HunyuanVideo for?

Researchers, developers and studios who want to run or build on a large open video model on their own GPUs.

HunyuanVideo at a glance

  • Open source: Source-available, not OSI open source: the Tencent Hunyuan Community License
  • GitHub: Tencent-Hunyuan/HunyuanVideo · 12,643 stars (as of Oct 21, 2026)

Key features

  • Open 13B-parameter text-to-video model
  • Weights, inference and parallel inference code
  • Diffusers integration and FP8 weights
  • Related I2V, Custom and Avatar models, plus HunyuanVideo-1.5
  • Penguin Video Benchmark released

HunyuanVideo FAQ

What is HunyuanVideo?

HunyuanVideo is Tencent's open video foundation model and framework for text-to-video generation, released with PyTorch model code, pre-trained weights and inference code. At release Tencent described it as the largest open video generation model, with over 13 billion parameters, and reported quality comparable to leading closed models.

Who is HunyuanVideo for?

Researchers, developers and studios who want to run or build on a large open video model on their own GPUs.

Is HunyuanVideo free?

HunyuanVideo's pricing, as the maker states it: Open weights (Tencent Hunyuan Community License; not valid in the EU, UK and South Korea). Check the official site for current plans.

Is HunyuanVideo open source?

No. HunyuanVideo's code is published under the Tencent Hunyuan Community License, which is not an OSI-approved open-source licence.

Can I use HunyuanVideo commercially?

It is released under the Tencent Hunyuan Community License, which does not apply in the European Union, United Kingdom and South Korea; read the licence for its terms before commercial use.

What are alternatives to HunyuanVideo?

Similar tools include ComfyUI, Wan 2.2, Mochi, Runway, Luma Dream Machine and LTX-2.

HunyuanVideo ratings and reviews

No ratings yet. Be the first to review HunyuanVideo.

Be the first to rate HunyuanVideo: tap 1 to 5 stars below. A written review is optional.

Your rating of HunyuanVideo

Star ratings count right away; written reviews appear after a quick check. One rating per tool per visitor. Reviews are from site visitors, not verified users. How reviews work

HunyuanVideo alternatives

Other video generation tools in our directory. See all video generation tools →

Related lists

🏆 See also: ComfyUI alternatives · All best-of lists

Badge for makers

Built HunyuanVideo? Add this badge to your README or site. It links back to this page. About badges

Listed on AgentScout (light)
Listed on AgentScout (dark)

Get the best new AI agents, weekly

One short email with new tools, launches and what's worth trying. Free, unsubscribe anytime.