fangwei123456/spikingjelly

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

2110 stars · 319 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

SpikingJelly is an open-source deep learning framework for Spiking Neural Networks (SNNs) built on PyTorch. It provides a beginner-friendly API for defining, training, and deploying SNNs, including ANN-to-SNN conversion, event-based datasets, and acceleration backends like torch, cupy, and triton.

Use cases

• build and train spiking neural networks with pytorch
• convert trained ANNs to energy-efficient SNNs
• train SNNs on event-based vision datasets like DVS
• simulate LIF neurons with surrogate gradient learning
• deploy spiking neural networks to neuromorphic hardware
• accelerate SNN training with cupy or triton backends
• run large-scale distributed SNN training

When to choose

• you want to experiment with spiking neural networks in a PyTorch-native way
• you need ANN-to-SNN conversion for energy-efficient inference
• you work with event-based camera datasets (DVS, N-MNIST, CIFAR10-DVS)
• you need a mature, documented SNN framework with active development

When to avoid

• you need conventional deep learning without spiking neurons
• you require a framework outside the PyTorch ecosystem
• you need production deployment on non-neuromorphic hardware without SNN support

Facets

framework · maturity active

deep-learning machine-learning simulation deep-learning machine-learning computer-vision python cross-platform spiking-neural-networks snn pytorch ann2snn neuromorphic event-based-vision dvs surrogate-gradients neuromorphic-computing gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="fangwei123456/spikingjelly")