Project AURIONAI Use-Cases Research & Innovation
Microsoft AI
vision

BEiT v2

BEiT with vector-quantised visual tokenisers, from the paper "BEiT v2".

Microsoft AIMicrosoft AIUSABEiTFamily
Figure from the BEiT v2 paper: ImageNet accuracy versus pretraining epochs for BEiT v2 against other self-supervised methods.
Levelvariant
Classvision
Parametersnot in the notes
Contextnot in the notes
Released12 Aug 2022
Opennessunknown
Runs onnot in the notes
Statusnot in the notes

Lineage

successor
BEiT-B to BEiT v2: successor, high confidenceBEiT v2 to BEiT-3: successor, high confidenceBEiT-BBEiT-BMicrosoft AI, Jun 2021BEiT v2BEiT v2Microsoft AI, Aug 2022BEiT-3BEiT-3Microsoft AI, Aug 2022
Microsoft AIBEiT-B is the parent (successor)high confidencenotes
Microsoft AIBEiT-3 is a child (successor)high confidencenotes
Where this comes from
Parent and child relations from model cards, Hugging Face base_model fields and the genealogy tables of the research notes. The tree shows two generations each way, 3 models. Confidence and evidence are the notes' own.

R&D Topics

Self-supervised visual representation learningproduced by

Identity

Developer
Microsoft AI
Family
BEiT
Released
12 Aug 2022 sourcesannouncement page
Weights published
–
Record checked
12 Oct 2026
Where this comes from
Developer and dates from the research notes. Release date is the vendor's public announcement; weights date is when open weights were published.

Architecture and Size

Architecture
dense transformer
Total parameters
– sources
Active parameters
–
Checkpoint parameters
–
Context length
not in the notes
Input
image
Output
labels and boxes
Where this comes from
Parameter counts as the developer states them. Where the checkpoint holds a different total, both are shown (register A16).

Openness and Access

Openness
unknown
Licence
not in the notes sources
Access routes
not in the notes
Where this comes from
The licence file governs over registry tags and aggregator labels.

Deployment

Runs on
not in the notes
Footprint
not in the notes
Runtimes
not in the notes
Where this comes from
Deployment tiers and runtimes as named in the notes. Footprint is the notes' figure, not a measurement by AURION.

Indonesia

Indonesian support
not stated
Regional languages
not in the notes
Jakarta hosting
unknown
Languages
not in the notes
Where this comes from
Indonesia fields are carried on every model; unknown is an allowed value.

Benchmark Results

The notes record no benchmark result for this model.

Industry Use Cases

No industry use case links this model.

Sources

  1. arxiv.org/abs/2208.06366 https://arxiv.org/abs/2208.06366

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "BEiT v2", https://aurion.semai.id/models/microsoft-beit-v2/, records of October 2026.