BindsNET/bindsnet

Health v2 · maintenance only

Adoption not part of the score

1695 stars · 347 forks observed · 2026-08-28

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

BindsNET is a Python package for simulating spiking neural networks (SNNs) built on PyTorch tensor functionality, running on CPUs or GPUs. It targets research on biologically inspired machine learning algorithms, including STDP-based learning and reinforcement learning applications.

Use cases

• simulate spiking neural networks in Python
• train SNNs with STDP learning rules
• run neuromorphic-style experiments on GPU with PyTorch
• apply spiking networks to reinforcement learning environments
• prototype biologically inspired neural network algorithms
• convert datasets into spike encodings for SNN experiments

When to choose

• you need a PyTorch-based SNN simulator with GPU acceleration
• you are researching biologically plausible learning rules like STDP
• you want to experiment with spiking networks on RL tasks
• you need a flexible, code-first SNN framework for research

When to avoid

• you need production deep learning with conventional ANNs
• you want a high-level GUI or no-code SNN tool
• you target neuromorphic hardware deployment rather than simulation
• you need a stable, long-term-supported production library

Facets

library · maturity active

machine-learning simulation deep-learning reinforcement-learning gpu-computing machine-learning deep-learning simulation python cross-platform spiking-neural-networks snn pytorch neuromorphic stdp biologically-inspired neuroscience gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="BindsNET/bindsnet")