Project AURIONAI Use-Cases Research & Innovation
Meta
vision

LeViT

Hybrid convolution and transformer classifier designed for fast inference, from the paper "LeViT: a Vision Transformer in ConvNet's Clothing".

MetaMetaUSA
Speed versus ImageNet accuracy plots of LeViT against EfficientNet and DeiT on one CPU thread and on a GPU, from the LeViT README.
Levelmodel
Classvision
Parametersnot in the notes
Contextnot in the notes
Released2 Apr 2021
Opennessopen source
Runs onmobile or NPU, CPU server
Statusactive

R&D Topics

Efficient vision networks for mobile and edge devicesproduced by

Identity

Developer
Meta
Family
not in the notes
Released
2 Apr 2021 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
hybrid
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
open source
Licence
Apache License 2.0 sources
Access routes
open weights download
Where this comes from
The licence file governs over registry tags and aggregator labels.

Deployment

Runs on
mobile or NPU, CPU server
Footprint
Edge or toolkit builds listed in: Qualcomm AI Hub, OpenVINO.
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.

Lineage

The notes record no lineage for this model.

Industry Use Cases

No industry use case links this model.

Sources

  1. arxiv.org/abs/2104.01136 https://arxiv.org/abs/2104.01136
  2. github.com/facebookresearch/LeViT https://github.com/facebookresearch/LeViT
  3. aihub.qualcomm.com/models/levit https://aihub.qualcomm.com/models/levit
  4. github.com/openvinotoolkit/open_model_zoo/tree/master/mo... https://github.com/openvinotoolkit/open_model_zoo/tree/master/models/public/levit-128s
  5. github.com/open-mmlab/mmpretrain/tree/main/configs/levit https://github.com/open-mmlab/mmpretrain/tree/main/configs/levit
  6. github.com/huggingface/pytorch-image-models/blob/main/RE... https://github.com/huggingface/pytorch-image-models/blob/main/README.md
  7. github.com/onnx/models/tree/main/Computer_Vision/levit_1... https://github.com/onnx/models/tree/main/Computer_Vision/levit_128s_Opset16_timm

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "LeViT", https://aurion.semai.id/models/meta-levit/, records of October 2026.