Mage-VL

Home/Models/microsoft/Mage-VL

microsoft/Mage-VL

kind: original
quant: bf16 (16-bit)
family: misc
modality: image-text-to-text
root model: microsoft/Mage-VL
provider: microsoft
license: apache-2.0
language: —
context: —
library: transformers
vRAM (GB): —
size on disk: —
downloads: 308
likes: 28
created: —
last modified: 2026-07-27T15:10:49+00:00 (2026)
huggingface: https://huggingface.co/microsoft/Mage-VL
tags: transformerssafetensorsmage_vlimage-text-to-textmultimodalvision-language-modelmage-vlvideo-understandingstreamingconversationalcustom_codelicense:apache-2.0region:us

benchmark	category	score	context	conf	evidence
DocVQA	multimodal	95.14	DocVQA-val	medium	`\| DocVQA-val \| 95.14 \| 94.69 \| 92.79 \| 76.20 \|`
DocVQA	multimodal	87.46	MultiDocVQA-val	medium	`\| MultiDocVQA-val \| 87.46 \| 87.21 \| 46.84 \| 58.35 \|`
ChartQA	multimodal	84.88	ChartQA	medium	`\| ChartQA \| 84.88 \| 83.96 \| 83.76 \| 83.40 \|`
MMBench	multimodal	84.02	MMBench-EN-dev	medium	`\| MMBench-EN-dev \| 84.02 \| 83.25 \| 65.81 \| 84.19 \|`
MMBench	multimodal	82.04	MMBench-CN-dev	medium	`\| MMBench-CN-dev \| 82.04 \| 80.58 \| 75.17 \| 79.47 \|`
ChartQA	multimodal	32.57	ChartQAPro	medium	`\| ChartQAPro \| 32.57 \| 26.79 \| 0.13 \| 25.38 \|`

none scraped.