TextImage + textVoice assistantsComputer-use agents

Muse Spark Open

Meta's largest open model: 816B parameters, 42B active per token, a 1 million token context. It understands text, images, video and audio and answers in text. Built for programming and long agentic tasks. Trained on data in more than 100 languages. Access to the weights on Hugging Face is granted manually.

Developer
Meta Superintelligence Labs, USA
First release
Oct 2026
Latest release
Oct 2026
Sizes
816B-A42B
License
Commercial use with conditionsMuse Model License: you may use, fine-tune and sell, but a "Built with Muse Spark Open" notice is required in the product and the usage policy must be followed; if you sell access to AI models by API (Model as a Service) and your revenue over 12 months is above 20 million dollars, a separate agreement with Meta is needed
Russian
Not stated
Running
On your own serverNeeds a GPU
Industries
Software development, Documents and accounting, Customer support, Science and research

What it does

  • an agent for working with code and long tasks with tool calling
  • analyzing screenshots, charts, documents, video and recordings
  • generating training data for other models
  • judging answers of other models (LLM as a judge)

Where it is used

software development and ITlarge companies with their own serversresearch groups

Hardware requirements

LaptopLaptop or regular PC, up to 8 GB of VRAM — smaller versions
no versions
1 GPUOne GPU with 16–80 GB — mid-size versions
no versions
ClusterServer with several GPUs — flagship versions
fits

Versions

  1. Muse Spark Open NVFP4 (для 700 ГБ видеопамяти)
  2. Muse Spark Open 816B-A42B (BF16)

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 Muse Spark Open be used in a commercial project?

With conditions. License: Muse Model License: you may use, fine-tune and sell, but a "Built with Muse Spark Open" notice is required in the product and the usage policy must be followed; if you sell access to AI models by API (Model as a Service) and your revenue over 12 months is above 20 million dollars, a separate agreement with Meta is needed. Restrictions vary — region, company revenue, attribution requirements. Have a lawyer check the terms before a commercial launch.

What hardware does Muse Spark Open need?

At minimum: Server with several GPUs — flagship 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 Muse Spark Open support Russian?

The model card does not list languages, so Russian support cannot be promised — it has to be tested on your own examples.

Where can I download Muse Spark Open and what does it cost?

The Muse Spark Open 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

Source: huggingface.co/meta-models/Muse-Spark-Open-816B-A42B. Data checked against the model card on 22 Sep 2026. Have a lawyer review the license before commercial launch.