The publication AI reads — a remote MCP server, nothing to install.Connect your agent →
Model register·microsoft

microsoft/Mage-VL-Cowork

microsoft/Mage-VL-Cowork is a microsoft record on this register. The Hugging Face model API lists the pipeline tag as image-text-to-text. The safetensors index gives an exact parameter count of 4,741,793,792. Architectural context length (max_position_embeddings) is 262,144. The listed licence is apache-2.0. The Hub repository was created on 30 Oct 2026, which is the release date this register uses.

Primary source: the Hugging Face model card for microsoft/Mage-VL-Cowork. Fields below come from that card, the Hub model API, and the repository’s own config.json when those files exist. A field we cannot source is omitted rather than guessed.

Organisationmicrosoft
ReleasedRepo creation date on Hugging Face30 Oct 2026
Last modified30 Oct 2026
ParametersExact, from the safetensors index4.7B 4,741,793,792
Context lengthmax_position_embeddings — architectural maximum, not necessarily served256K 262,144
Licenceapache-2.0
Pipeline / taskpipeline_tag on the Hugging Face model APIimage-text-to-text
Base modelDeclared on the Hub; not a page on this registermicrosoft/Mage-VL
File formatssafetensors
Librarylibrary_name on the Hugging Face model APItransformers
Architecturellava_onevision2
Layers36
Vocabulary151,936
DownloadsAt last check1.5k
Likes152
GatedNo

Lineage

The Hub card names microsoft/Mage-VL as the base model. That id is not a page on this register.

How to run

Safetensors weights are on the Hugging Face repository. The parameter count on this page came from that index.

The Hugging Face model API lists the library as transformers.

Load it from huggingface.co/microsoft/Mage-VL-Cowork. This page does not add install steps that the Hub files do not state.

Agent capabilities

Derived from the model’s own chat template, with the match counts behind each finding.

  • ●Tool calling23 matches in template
  • ●Reasoning / thinking16 matches in template
  • ●System prompt3 matches in template
  • ●Image input4 matches in template
  • ●Video input4 matches in template
  • ○Audio inputno evidence in template

Questions this record can answer

When was Mage-VL-Cowork released?
The Hugging Face repository was created on 30 Oct 2026, which is the date this register uses.
How many parameters does Mage-VL-Cowork have?
The safetensors index gives 4,741,793,792 parameters, shown here as 4.7B. That figure is exact, not rounded.
What is the context length of Mage-VL-Cowork?
262,144 tokens. That is max_position_embeddings from the model's own config.json — the architectural maximum, not necessarily the context a provider serves.
What licence is Mage-VL-Cowork under?
The Hub card lists apache-2.0.
Is Mage-VL-Cowork gated?
No. The Hugging Face model API does not mark this repository as gated.
Does the Mage-VL-Cowork chat template show tool calling?
The chat template contains evidence of tool calling.

Where each field came from

  • Hugging Face model API — releasedAt, lastModified, downloads, likes, task, license, gated, parameters, libraryName, formats, baseModel. Checked 3 Nov 2026.
  • config.json — contextLength, layers, vocabSize, architecture. Checked 3 Nov 2026.
  • chat template — capabilities. Checked 3 Nov 2026.

First recorded 1 Nov 2026, last checked 3 Nov 2026. This record is compiled automatically from the sources above. Where a field is blank, we could not source it and have not guessed. Canonical repository: huggingface.co/microsoft/Mage-VL-Cowork.