NVlabs/MambaVision
A hybrid vision backbone combining Mamba state-space models with self-attention for image classification, object detection, instance segmentation, and semantic segmentation.
Not currently ranked — collecting fresh signals.
star history
MambaVision is a hierarchical vision architecture that achieves state-of-the-art accuracy and throughput by combining selective state space models (Mamba) with transformer self-attention blocks. Developed by NVIDIA Research, it serves as a universal vision backbone supporting multiple downstream tasks including image classification, object detection, instance segmentation, and semantic segmentation. The model is implemented in PyTorch with support for Hugging Face integration.
Frequently asked
- What is NVlabs/MambaVision?
- A hybrid vision backbone combining Mamba state-space models with self-attention for image classification, object detection, instance segmentation, and semantic segmentation.
- Is MambaVision open source?
- Yes — NVlabs/MambaVision is an open-source project tracked on heatdrop.
- What language is MambaVision written in?
- NVlabs/MambaVision is primarily written in Python.
- How popular is MambaVision?
- NVlabs/MambaVision has 2.2k stars on GitHub.
- Where can I find MambaVision?
- NVlabs/MambaVision is on GitHub at https://github.com/NVlabs/MambaVision.