The Rundown AI homepage

Independent tool overview

Hunyuan Motion 1.0 at a glance

Tencent's downloadable text-to-motion model for generating skeleton-based 3D human animations from written action prompts.

Visit the official Hunyuan Motion 1.0 site ↗
Hunyuan Motion 1.0 product preview
Developer
Tencent Hunyuan 3D Digital Human Team
Primary use
Text-to-3D human motion generation
Model sizes
1.0B standard and 0.46B Lite
Minimum listed VRAM
26GB standard; 24GB Lite
Platforms
macOS, Windows, and Linux
License
Tencent HY-MOTION 1.0 Community License

Overview

What Hunyuan Motion 1.0 is

Hunyuan Motion 1.0, officially named HY-Motion 1.0, is Tencent's text-to-motion model family for generating skeleton-based 3D human animation from natural-language action descriptions. It is aimed more at developers and technical 3D teams than at creators looking for a polished, subscription-based animation app.

Tencent provides a 1-billion-parameter standard model and a 460-million-parameter Lite model, along with inference code, downloadable weights, command-line batch generation, and a local Gradio interface. The official workflow can produce an interactive motion preview and, when Autodesk's FBX SDK is available, downloadable FBX files for downstream animation work.

The weights have no download fee, but running the model is not cost-free: the standard setup calls for substantial GPU memory, local installation, storage, and engineering time. The custom Tencent community license also has geographic and commercial restrictions that teams should review before adopting it.

Use cases

Who Hunyuan Motion 1.0 is best for

The strongest fit depends on the job you need the product to complete, not the size of its feature list.

Technical animation teams

Generate human motion candidates from written action descriptions for use in a broader 3D production pipeline.

Game and virtual-character prototyping

Explore new humanoid actions before committing to motion capture or hand-keyed animation.

Researchers

Study a billion-parameter flow-matching approach to text-conditioned human motion generation using published weights and code.

Batch motion generation

Run collections of text or JSON prompts locally through the included command-line inference workflow.

Capabilities

Core Hunyuan Motion 1.0 features

1

Text-to-motion generation

Creates skeleton-based human movement from prompts describing actions and detailed limb or torso motion.

2

Two model sizes

Offers the full 1.0B model and a 0.46B Lite option, although both still have demanding official VRAM requirements.

3

Local batch inference

Processes directories of text or JSON prompt files through the included command-line script.

4

Local visual interface

Includes a Gradio app for entering prompts, adjusting duration and guidance, and previewing generated motion.

5

Prompt rewriting and duration estimation

Supports an optional language-model module that can rewrite prompts and estimate an appropriate clip duration.

6

FBX export path

The official app can expose FBX downloads when the Autodesk FBX SDK dependency is available in the environment.

Process

How the Hunyuan Motion 1.0 workflow works

  1. Step 1

    Review the license

    Confirm that your location, use case, distribution plan, and organization size fit Tencent's community-license terms.

  2. Step 2

    Install the project

    Clone the official repository, install PyTorch and the required Python packages, and pull the large model files.

  3. Step 3

    Choose a checkpoint

    Select the 1.0B standard model or the smaller Lite model based on output needs and available GPU memory.

  4. Step 4

    Write an action-focused prompt

    Describe one humanoid subject's body movement in English and keep the wording concise for the most reliable results.

  5. Step 5

    Generate and integrate

    Run the prompt through the CLI or local interface, inspect the motion, then export or retarget it for the destination character pipeline.

Cost

Hunyuan Motion 1.0 pricing and free plan

Tencent does not charge a subscription or model-download fee for HY-Motion 1.0. Users supply their own computing, storage, setup, and maintenance, and usage is governed by a custom community license rather than a standard permissive open-source license.

Community model

No download fee

Inference code and model weights are publicly downloadable for eligible users under Tencent's HY-MOTION 1.0 Community License.

  • Self-hosted compute and storage costs are separate
  • The license excludes the EU, UK, and South Korea
  • Organizations above the license's stated 1 million-MAU threshold must request separate permission
  • Distribution, acceptable-use, and attribution conditions apply

Pricing checked . Check current pricing at the source ↗

Assessment

Hunyuan Motion 1.0 strengths and limitations

Where it stands out

  • Produces motion data rather than only a rendered video
  • Includes standard and Lite checkpoints
  • Supports both batch inference and a local visual interface
  • Official documentation lists Windows, macOS, and Linux support
  • The repository includes prompting guidance and evaluation resources

What to consider

  • The official minimum is still 24GB of VRAM for Lite and 26GB for the standard model
  • Setup requires Python, PyTorch, large model downloads, and familiarity with local inference
  • It is designed for single humanoid motion, not animals, non-human creatures, or multi-person interactions
  • Prompts about appearance, clothing, environments, objects, and camera direction are outside the model's stated focus
  • Seamless loops and in-place animation are not officially supported
  • The custom license excludes use in the EU, UK, and South Korea and adds conditions for some commercial deployments

Compare

Hunyuan Motion 1.0 alternatives

The right alternative depends on the specific output, workflow, controls and budget your project requires.

Design

Move AI Genesis

Choose Move AI when you want to capture human motion from video instead of generating movement from text.

Explore Move AI Genesis

Design

Rodin

Choose Rodin when the main need is generating the 3D character or object asset rather than its skeletal motion.

Explore Rodin

Content Creator

Runway Creative

Choose Runway when the intended result is a finished generative video rather than reusable 3D animation data.

Explore Runway Creative

Questions

Hunyuan Motion 1.0 FAQs

What is Hunyuan Motion 1.0?

Hunyuan Motion 1.0, or HY-Motion 1.0, is Tencent's model family for turning written action prompts into skeleton-based 3D human motion.

Is Hunyuan Motion 1.0 free?

The official code and weights have no download fee, but you pay for the hardware or cloud compute used to run them. Usage is also subject to Tencent's custom community license.

How much GPU memory does HY-Motion need?

Tencent lists a 26GB minimum for the 1.0B standard model and 24GB for the 0.46B Lite model. The optional prompt-engineering module can require additional memory.

Can HY-Motion export FBX files?

The official local interface includes an FBX download path when the Autodesk FBX SDK is available. Teams should still expect retargeting and cleanup before using the motion on a production character.

Does HY-Motion support animals or multiple characters?

No. Tencent's current guidance says the model does not support non-humanoid characters or multi-person interactions.

Can companies use HY-Motion commercially?

The community license permits eligible uses but includes geographic, distribution, acceptable-use, and scale restrictions. It excludes the EU, UK, and South Korea, and organizations above the stated 1 million-MAU threshold must request a separate license. Legal review is appropriate before production use.

Bottom line

Our Hunyuan Motion 1.0 verdict

HY-Motion 1.0 is a compelling technical option for teams that need locally generated, reusable human motion data and have high-memory GPUs plus a retargeting pipeline. It is a poor fit for casual users, restricted regions, or anyone expecting a turnkey hosted animation product.

Visit Hunyuan Motion 1.0 website ↗
The Rundown University

AI training for the future of work.

Get access to all our AI courses, hundreds of real-world AI use cases, live expert-led workshops, an exclusive network of AI early adopters, and more.

AI Courses

Get unlimited access to all of our current & upcoming industry-specific AI courses for the duration of your subscription.

Daily Guides

To keep up with the rapid pace of AI, our team publishes AI implementation guides daily. Our library contains 300+ practical use cases to automate real-world work.

Workshops

Join weekly, live, interactive sessions with industry leaders who are at the forefront of AI for hands-on implementation guidance and exclusive insights.

Community

Network with an exclusive community of AI-first professionals who are working smarter with AI. Learn how early adopters are using AI in their work and businesses.