NEWVectors or files. Pick a path.Start →
    Models/Image Text To Text/microsoft/Mage-VL
    Image Text To Texttransformersapache-2.0

    Mage-VL

    by microsoft

    Vision-language model built for streaming video rather than single frames

    Identifier
    Model ID
    microsoft/Mage-VL

    Overview

    Most vision-language models take one image and some text. Mage-VL is tagged for video understanding and streaming, which is a different problem: the input keeps arriving, and the model has to stay useful without re-reading everything it has already seen.

    That matters for anyone indexing video, because the naive approach is to sample frames, embed each one independently, and lose every relationship between them. A model that consumes a stream can describe what changed, not just what is present.

    At 4.7B parameters under Apache 2.0 it sits in the range you can self-host. Treat it as a describer and a reranker rather than a first-pass indexer: generating text for every frame of a large archive is expensive, and a cheap embedding model is the right thing to run first.

    Architecture

    MageVLForConditionalGeneration, model type mage_vl, 4,741,793,792 parameters. Image-text-to-text pipeline with video-understanding and streaming support declared on the model card. Requires trust_remote_code.

    Key Capabilities

    • Describing video segments in natural language
    • Question answering grounded in visual content
    • Streaming input rather than fixed-length clips
    • Generating text metadata for otherwise unlabelled footage

    Use Cases on Mixpeek

    • Writing searchable descriptions for archive footage that has no captions
    • Answering questions about a video without a human watching it
    • Producing the text side of a hybrid index, where BM25 needs something to match
    • Reranking a shortlist of clips returned by a cheaper embedding model

    Tags

    transformerssafetensorsmage_vlimage-text-to-textmultimodalvision-language-modelmage-vlvideo-understandingstreamingconversationalcustom_codearxiv:2607.24904license:apache-2.0region:us

    Use Mage-VL on Mixpeek

    Build multimodal processing pipelines with this model and others. Extract features, run inference, and set up retrieval in Mixpeek Studio.

    Open Studio

    How It Runs on Mixpeek

    On Mixpeek, Mage-VL runs as a managed extractor inside a processing pipeline. Point a bucket of image text to text data at it, and Mixpeek handles GPU provisioning, batching, retries, and writing the outputs into a vector store you can query.

    Extractor outputs land in the Mixpeek Vector Store (MVS), where you can combine them with retrieval, reranking, and filter stages to build end-to-end search and agent-perception pipelines, no model-serving infrastructure to maintain.