Project AURIONAI Use-Cases Research & Innovation
MIT Han Lab

MIT Han Lab

University, USA. 6 models in the directory.

universityUSA
0deployments

Identity

Type
university
Country label
USA
HQ country
USA
Founding country
USA
Parent
not in the notes
Parent country
not in the notes
Country basis
the research notes
Where this comes from
Convention D1: a dual label, HQ first, when the HQ, founding or parent country differ.

Models by Class

Models

ModelClassOpennessParametersReleased
MIT Han LabEfficientViT-SAMvisionopen weights permissive–7 Feb 2024MIT Han LabOnce-for-All (OFA)visionopen source–26 Aug 2019MIT Han LabProxylessNASvisionopen source–2 Dec 2018MIT Han LabMCUNetTinyMLopen source––MIT Han Labmcunet-in4TinyMLopen source––MIT Han Labmcunet-vww2TinyMLopen source––

Scores

BenchmarkModelScoreWho ran itSetting
ImageNet top-1mcunet-in468 %vendor-reportedint8, top-1
Visual Wake Words (VWW)mcunet-vww291.8 %vendor-reportedint8
Where this comes from
Every recorded score of this organisation's models, with who ran it. Scores from different settings are not directly comparable.

Sources

  1. GitHub - mit-han-lab/mcunet: [NeurIPS 2020] MCUNet: Tiny Deep Learning on IoT Devices; [NeurIPS 2021] MCUNetV2: Memory-Efficient Patch-based github.com https://github.com/mit-han-lab/mcunet

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "MIT Han Lab", https://aurion.semai.id/organisations/mit-han-lab/, records of October 2026.