Unlike traditional models that lag in synchronizing multiple modalities, HuMo, ByteDance’s latest release, introduces a unified human-centric video generation (HCVG) framework capable of producing highly realistic, fine-grained, and controllable human videos. What makes HuMo stand out is its ability to seamlessly combine text, image, and audio inputs to generate videos that are not only visually compelling but also contextually consistent. Want to craft a character with a specific appearance, outfit, and setting? Simply pair text prompts with reference images. Or need lifelike lip-syncing and gestures to match audio narration or dialogue? HuMo achieves this with audio-driven motion synchronization and a novel focus-by-predicting strategy that aligns audio precisely with facial regions. Even better, if you want full creative control, you can combine all three, text, image, and audio, for the most flexible and professional results. Under the hood, HuMo uses a two-stage multimodal training paradigm with minimal-invasive image injection for subject preservation and a time-adaptive classifier-free guidance system for dynamic control. Practically speaking, this means your videos follow prompts accurately, preserve character identity across scenes, and maintain natural audio-visual sync, capabilities that surpass specialized SOTA methods while being packaged into a single unified framework.
This guide for walk you through the step-by-step process to get this model up and running in GPU-based environments and generate example video using text, image and audio input.
Prerequisites
The minimum system requirements for running this model are:
- GPU: 2x H100 or 1x H200
- Storage: 200 GB (preferable)
- VRAM: at least 110 GB
- Anaconda installed
Step-by-step process to install and run ByteDance HuMo
For the purpose of this tutorial, we’ll use a GPU-powered Virtual Machine by NodeShift since it provides high compute Virtual Machines at a very affordable cost on a scale that meets GDPR, SOC2, and ISO27001 requirements. Also, it offers an intuitive and user-friendly interface, making it easier for beginners to get started with Cloud deployments. However, feel free to use any cloud provider of your choice and follow the same steps for the rest of the tutorial.
Step 1: Setting up a NodeShift Account
Visit app.nodeshift.com and create an account by filling in basic details, or continue signing up with your Google/GitHub account.
If you already have an account, login straight to your dashboard.
Step 2: Create a GPU Node
After accessing your account, you should see a dashboard (see image), now:
- Navigate to the menu on the left side.
- Click on the GPU Nodes option.
- Click on Start to start creating your very first GPU node.
These GPU nodes are GPU-powered virtual machines by NodeShift. These nodes are highly customizable and let you control different environmental configurations for GPUs ranging from H100s to A100s, CPUs, RAM, and storage, according to your needs.
Step 3: Selecting configuration for GPU (model, region, storage)
- For this tutorial, we’ll be using 1x H200 GPU, however, you can choose any GPU as per the prerequisites.
- Similarly, we’ll opt for 200GB storage by sliding the bar. You can also select the region where you want your GPU to reside from the available ones.
Step 4: Choose GPU Configuration and Authentication method
- After selecting your required configuration options, you’ll see the available GPU nodes in your region and according to (or very close to) your configuration. In our case, we’ll choose a 1x H200 140GB GPU node with 192 vCPUs/252GB RAM/200GB SSD.
2. Next, you’ll need to select an authentication method. Two methods are available: Password and SSH Key. We recommend using SSH keys, as they are a more secure option. To create one, head over to our official documentation.
Step 5: Choose an Image
The final step is to choose an image for the VM, which in our case is Nvidia Cuda.
In our previous blogs, we used pre-built images from the Templates tab when creating a Virtual Machine. However, for running a CUDA dependent model like HunyuanImage-2.1, we need a latest CUDA version like CUDA12.4. Therefore, in this case, we switched to the Custom Image tab and selected a specific Docker image that meets all runtime and compatibility requirements.
We chose the following image:
nvidia/cuda:12.4.0-devel-ubuntu24.04
This image is essential because it includes:
- Full CUDA toolkit (including
nvcc)
- Proper support for building and running GPU-based applications
- Compatibility with CUDA 12.4.0 required by model operations
Launch Mode
We selected:
Interactive shell server
This gives us SSH access and full control over terminal operations — perfect for installing dependencies, running benchmarks, and launching models.
Docker Repository Authentication
We left all fields empty here.
Since the Docker image is publicly available on Docker Hub, no login credentials are required.
Identification
nvidia/cuda:12.4.0-devel-ubuntu24.04
That’s it! You are now ready to deploy the node. Finalize the configuration summary, and if it looks good, click Create to deploy the node.
Step 6: Connect to active Compute Node using SSH
- As soon as you create the node, it will be deployed in a few seconds or a minute. Once deployed, you will see a status Running in green, meaning that our Compute node is ready to use!
- Once your GPU shows this status, navigate to the three dots on the right, click on Connect with SSH, and copy the SSH details that appear.
As you copy the details, follow the below steps to connect to the running GPU VM via SSH:
- Open your terminal, paste the SSH command, and run it.
2. In some cases, your terminal may take your consent before connecting. Enter ‘yes’.
3. A prompt will request a password. Type the SSH password, and you should be connected.
Output:
Next, If you want to check the GPU details, run the following command in the terminal:
!nvidia-smi
Step 7: Set up the project environment with dependencies
- Create a virtual environment using Anaconda.
conda create -n humo python=3.11 -y && conda activate humo
Output:
2. Clone the official repository of the project.
https://github.com/Phantom-video/HuMo.git && cd HuMo
3. Install required dependencies.
pip install torch==2.5.1 torchvision==0.20.1 torchaudio==2.5.1 --index-url https://download.pytorch.org/whl/cu124
pip install flash_attn --no-build-isolation
pip install -r requirements.txt
conda install -c conda-forge ffmpeg
4. Login to Hugging Face CLI.
huggingface-cli login
5. Edit the following scripts with nano and replace the value of GPU nodes from “8” to “1” or to the exact number of GPU nodes you are actually have in the VM.
nano scripts/infer_tia.sh
nano humo/configs/inference/generate.yaml
Here’s the lines you need to edit:
Step 8: Run the model for inference
- Download the checkpoints of all the models.
huggingface-cli download Wan-AI/Wan2.1-T2V-1.3B --local-dir ./weights/Wan2.1-T2V-1.3B
huggingface-cli download bytedance-research/HuMo --local-dir ./weights/HuMo
huggingface-cli download openai/whisper-large-v3 --local-dir ./weights/whisper-large-v3
Output:
2. Run the model for inference by customizing your prompt in the ./examples/text_case.json file as shown here.
bash scripts/infer_tia.sh
Please note if you’re using less number of GPU nodes, the video generation is going to take longer, almost around 40-50 minutes.
Output:
Here’s the final generated video:
If video has failed to preview, here’s an alternate link: https://drive.google.com/drive/folders/1T93J27ihVMJIzrlX-uNsC4eO8PThTDvD?usp=sharing
Conclusion
HuMo’s human-centric video generation by unifying text, image, and audio into a single, finely tuned framework delivers precision, realism, and unmatched creative control. At the same time, running and experimenting with HuMo becomes seamless with NodeShift, which takes care of the heavy lifting by providing the compute, scalability, and deployment-ready infrastructure needed to unlock its full potential. Together, HuMo’s groundbreaking capabilities and NodeShift Cloud’s effortless setup create a perfect combination for researchers, developers, and creators eager to push the boundaries of next-gen video generation.
