What are the hardware requirements for Mage VL?

Mage VL requires approximately 9.5 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Mage VL and how large is it?

Mage VL is developed by microsoft, a model with 4.7b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Mage VL?

Mage VL is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Mage VL by microsoft — VRAM 9.5GB

Name: Mage VL
Author: microsoft

Mage VL is an open-source language model by microsoft. Features: 4.7b LLM, VRAM: 9.5GB, License: apache-2.0, LLM Explorer Score: 0.52.

Arxiv:2607.24904 Conversational Custom code Image-text-to-text Mage-vl Mage vl Multimodal Region:us Safetensors Sharded Streaming Tensorflow Video-understanding Vision Vision-language-model

Model Card on HF 🤗: https://huggingface.co/microsoft/Mage-VL

Mage VL Parameters and Internals

LLM Name	Mage VL
Repository 🤗	https://huggingface.co/microsoft/Mage-VL
Model Size	4.7b
Required VRAM	9.5 GB
Updated	2026-08-08
Maintainer	microsoft
Model Type	mage_vl
Model Files	5.0 GB: 1-of-2 4.5 GB: 2-of-2 1.1 GB
Model Architecture	MageVLForConditionalGeneration
License	apache-2.0
Model Max Length	131072
Transformers Version	5.7.0
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Errors	replace

Rank the Mage VL capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation