MzeroMiko/VMamba

Health v2 · maintenance only

Adoption not part of the score

3219 stars · 237 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

VMamba is a PyTorch implementation of a visual state space model (SSM) vision backbone based on Mamba, featuring 2D Selective Scan (SS2D) for linear-complexity image feature extraction. It provides pretrained model weights and a single-file implementation for easy use in downstream vision tasks.

Use cases

• build image classification models with linear-complexity backbones
• use state space models for computer vision tasks
• replace ViT or CNN backbones with Mamba-based vision models
• run object detection or segmentation with a VMamba backbone
• experiment with 2D selective scan architectures

When to choose

• you need an efficient vision backbone with linear time complexity
• you want to experiment with state space models in vision
• you need pretrained weights for downstream vision tasks

When to avoid

• you need a battle-tested production backbone with broad ecosystem support
• you lack a CUDA-capable GPU since mamba_ssm requires GPU kernels
• you only need standard CNN or transformer architectures

Facets

library · maturity active

machine-learning deep-learning computer-vision deep-learning machine-learning python state-space-models vision-backbone mamba image-classification neural-architecture linux gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="MzeroMiko/VMamba")