Task
Video Classification
V
qubvel-hf/vjepa2-vitl-fpc16-256-ssv2
Video Classification·qubvel-hf· 58· 7
transformers
mit
Video Classification
375.5M params
dataset:HuggingFaceM4/something_something_v2base_model:facebook/vjepa2-vitl-fpc64-256base_model:finetune:facebook/vjepa2-vitl-fpc64-256license:mitregion:us
A frontier video understanding model developed by FAIR, Meta, which extends the pretraining objectives of VJEPA, resulting in state-of-the-art video understanding capabilities, leveraging data and model sizes at scale. The code is released in this repository.
# pull & run locally
pip install mlforge-sdk && mlforge pull qubvel-hf/vjepa2-vitl-fpc16-256-ssv2
pip install mlforge-sdk && mlforge pull qubvel-hf/vjepa2-vitl-fpc16-256-ssv2
Model details
Provider
qubvel-hf
Framework
transformers
Parameters
375.5M
Size
1.4 GB
License
mit
Downloads
58
Likes
7
Updated
2025-06-13
About qubvel-hf/vjepa2-vitl-fpc16-256-ssv2
A frontier video understanding model developed by FAIR, Meta, which extends the pretraining objectives of VJEPA, resulting in state-of-the-art video understanding capabilities, leveraging data and model sizes at scale. The code is released in this repository.