Image + text

VideoLLaMA

Models that watch a video and answer questions about it: what happens, when, who does what. VideoLLaMA 3 at 2B and 7B is among the strongest in its size class.

Developer
Alibaba DAMO Academy, China
First release
May 2023
Latest release
Jan 2025
Sizes
2B – 72B
License
Commercial use allowedApache 2.0 (VideoLLaMA 2 and 3); the first version is built on LLaMA and carries its restrictions
Russian
Not supported
Running
On your own serverNeeds a GPU
Industries
Media and production, Security, Education
Deploy on your server Hugging Face GitHub

What it does

  • Video description and short summary
  • Finding a moment in a recording by question
  • Tagging a video archive
  • Reviewing videos before publishing

Where it is used

Media and video productionVideo surveillanceTraining and online courses

Hardware requirements

LaptopLaptop or regular PC, up to 8 GB of VRAM — smaller versions
fits
1 GPUOne GPU with 16–80 GB — mid-size versions
fits
ClusterServer with several GPUs — flagship versions
fits

Versions

  1. VideoLLaMA 3 (2B, 7B)
  2. VideoLLaMA 2.1 (в т.ч. аудио+видео)
  3. VideoLLaMA 2
  4. Video-LLaMA

How to run it

On your own serverWeights are downloaded from Hugging Face and served with vLLM (load and API) or llama.cpp (modest hardware). The model then runs in a closed network with no per-request fees.Hugging Face
How much hardware you needCalculate the VRAM for the model size, context length and number of concurrent requests.Open the hardware calculator

I can set this up end to end: pick the model size, deploy it on your server and connect it to your systems.

Frequently asked questions

Can VideoLLaMA be used in a commercial project?

Yes. License: Apache 2.0 (VideoLLaMA 2 and 3); the first version is built on LLaMA and carries its restrictions. It allows commercial use, but it is still worth having a lawyer review the license before launch.

What hardware does VideoLLaMA need?

At minimum: Laptop or regular PC, up to 8 GB of VRAM — smaller versions. Without a GPU the model is not practical. You can calculate the exact VRAM for your model size and context in the hardware calculator.

Does VideoLLaMA support Russian?

No. The model card lists its languages and Russian is not among them.

Where can I download VideoLLaMA and what does it cost?

The VideoLLaMA weights are open and free to download. You only pay for the hardware it runs on and for the setup. Source links are at the bottom of this page.

How I deploy it for clients

  1. SelectionI pick the model size for your task and hardware and test it on your examples.
  2. DeploymentI deploy it on your server or in a closed network and provide an API.
  3. Fine-tuningI fine-tune it on your data (LoRA) or connect a knowledge base — whichever is cheaper for the task.
  4. IntegrationI connect it to your CRM, ERP, bot, website or team chat and set up monitoring.

Similar models

Image + textInternVideoShanghai AI Lab (OpenGVLab) · ChinaCommercial use allowed

A family of video models: encoders for search and classification of clips, and chat models that analyze long videos. InternVideo 3 is designed for multi-hour recordings.

DetailsImage + textQwen-VLAlibaba (Qwen team) · ChinaCommercial use allowed

One of the strongest open vision models: reads documents, tables, charts and video, and works with user interfaces. Since Qwen3.5, vision is built directly into the main Qwen model.

DetailsImage + textLLaVALLaVA / LMMs-Lab (researchers from the USA and China) · USA / ChinaCommercial use allowed

The open project that started the trend for image-plus-text models. The OneVision line understands photos, documents and video; training data and recipes are open.

Details

Source: github.com/DAMO-NLP-SG/VideoLLaMA3. Data checked against the model card on 22 Sep 2026. Have a lawyer review the license before commercial launch.