mlx-snn

A general-purpose Spiking Neural Network library built on Apple MLX.

mlx-snn aims to provide an efficient, research-friendly SNN framework that leverages MLX's unified memory architecture and lazy evaluation. Whether you're exploring neuron dynamics, training classifiers with surrogate gradients, or exchanging models via NIR, mlx-snn offers a clean, Pythonic API that integrates naturally into the MLX ecosystem.

Why mlx-snn?

• MLX-native — All operations use mlx.core. No PyTorch/CUDA dependency. Runs on Apple Silicon with zero-copy unified memory.
• Research-friendly — Explicit state dicts, composable surrogate gradients, and standard mlx.nn.Module patterns make it easy to experiment and extend.
• Cross-framework — NIR support lets you import and export models to/from snnTorch, Norse, SpikingJelly, and neuromorphic hardware platforms.
• Hardware tested — Currently validated on Apple M3 Max. Future Apple Silicon releases will be tested and supported as they become available.

Installation

pip install mlx-snn

Requires Python 3.9+ and Apple Silicon (M1/M2/M3/M4).

Quick Start

import mlx.core as mx
import mlx.nn as nn
import mlxsnn

# Build a spiking network
fc = nn.Linear(784, 10)
lif = mlxsnn.Leaky(beta=0.95, threshold=1.0)

# Encode input as spike train and run over time
spikes_in = mlxsnn.rate_encode(mx.random.uniform(shape=(8, 784)), num_steps=25)
state = lif.init_state(batch_size=8, features=10)

for t in range(25):
    spk, state = lif(fc(spikes_in[t]), state)

print("Output membrane:", state["mem"].shape)  # (8, 10)

Features

Neuron Models

Model	Since	Description
Leaky (LIF)	v0.1	Leaky Integrate-and-Fire with configurable decay
IF	v0.1	Integrate-and-Fire (non-leaky)
Izhikevich	v0.2	2D dynamics with RS/IB/CH/FS presets
Adaptive LIF	v0.2	LIF with adaptive threshold
Synaptic	v0.2	Conductance-based dual-state LIF
Alpha	v0.2	Dual-exponential synaptic model
RLeaky	v0.4	Recurrent LIF with learnable feedback weight
RSynaptic	v0.4	Recurrent Synaptic with learnable feedback weight

Surrogate Gradients

All neuron models support differentiable training via surrogate gradients:

• Fast Sigmoid — default, good balance of speed and accuracy
• Arctan — smoother gradient landscape
• Sigmoid — standard logistic sigmoid derivative
• Triangular (Tent) — localized, compact support near threshold
• Straight-Through Estimator — simplest, unit gradient everywhere
• Custom — plug in any smooth approximation

Spike Encoding

Method	Since	Use Case
Rate (Poisson)	v0.1	Static images, general-purpose
Latency (TTFS)	v0.1	Energy-efficient, temporal coding
Delta Modulation	v0.2	Temporal signals, change detection
EEG Encoder	v0.2	EEG-to-spike with frequency band support

Training & Loss Functions

• BPTT forward pass helper (bptt_forward)
• Loss functions: ce_rate_loss, ce_count_loss, mse_membrane_loss, membrane_loss, rate_coding_loss
• Learnable parameters: learn_beta, learn_threshold, learn_V on all neurons
• Works with standard MLX optimizers (mlx.optimizers.Adam, etc.)

NIR Interoperability

NIR (Neuromorphic Intermediate Representation) enables cross-framework SNN model exchange between simulators and neuromorphic hardware platforms.

pip install mlx-snn[nir]

Export an mlx-snn model to NIR:

import mlx.nn as nn
import mlxsnn, nir

layers = [
    ('fc1', nn.Linear(784, 128)),
    ('lif1', mlxsnn.Leaky(beta=0.9)),
    ('fc2', nn.Linear(128, 10)),
    ('lif2', mlxsnn.Leaky(beta=0.9)),
]
graph = mlxsnn.export_to_nir(layers)
nir.write('model.nir', graph)

Import a NIR model into mlx-snn:

graph = nir.read('model.nir')
model = mlxsnn.import_from_nir(graph)
state = model.init_states(batch_size=32)
out, state = model(x, state)

Supported conversions: nn.Linear <-> nir.Affine/nir.Linear, Leaky <-> nir.LIF, IF <-> nir.IF, Synaptic <-> nir.CubaLIF.

Benchmark Highlights

Experiments on MNIST (784-128-10 SNN, 25 timesteps, 5 seeds) on Apple M3 Max, compared with snnTorch on NVIDIA V100:

Configuration	mlx-snn (M3 Max)	snnTorch (V100)	Speed (mlx-snn)	Speed (snnTorch)
Leaky (LIF)	96.3%	97.3%	5.7 s/epoch	20.9 s/epoch
Synaptic	94.4%	95.8%	6.1 s/epoch	25.2 s/epoch
RLeaky (V=0.1, learn)	91.6%	68.1%	6.8 s/epoch	25.7 s/epoch
RSynaptic (V=0.1, learn)	89.0%	52.2%	7.3 s/epoch	29.2 s/epoch
Fast Sigmoid surrogate	96.3%	96.7%	5.7 s/epoch	20.9 s/epoch
Triangular (Tent) surrogate	86.0%	50.8%	10.9 s/epoch	20.9 s/epoch

mlx-snn achieves ~3.7-4.1x faster training per epoch on the M3 Max compared to the V100, while maintaining competitive accuracy. Recurrent neurons with learnable weights significantly outperform snnTorch's default configurations.

For full results, see our benchmarking paper and the experiments/ directory.

Migrating from snnTorch

mlx-snn is designed to feel familiar to snnTorch users:

# snnTorch                          # mlx-snn
import snntorch as snn              import mlxsnn
lif = snn.Leaky(beta=0.9)          lif = mlxsnn.Leaky(beta=0.9)
spk, mem = lif(x, mem)             spk, state = lif(x, state)
                                    # state["mem"] == mem

Key differences:

• State is a dict, not separate tensors — plays well with MLX functional transforms
• No global hidden state — state is always explicit (pass in, get out)
• MLX arrays instead of PyTorch tensors — use mx.array, not torch.Tensor
• Surrogate gradients use the STE pattern with mx.stop_gradient

Project Structure

mlxsnn/
├── neurons/       # SpikingNeuron base, Leaky, IF, Izhikevich, ALIF, Synaptic, Alpha, RLeaky, RSynaptic
├── surrogate/     # fast_sigmoid, arctan, sigmoid, triangular, straight_through, custom
├── encoding/      # rate, latency, delta, EEG encoder
├── functional/    # Stateless pure functions (lif_step, if_step, fire, reset)
├── training/      # BPTT helpers, loss functions
└── nir_*.py       # NIR export/import utilities

Roadmap

• v0.1 — Core neurons (LIF, IF), surrogate gradients, rate/latency encoding
• v0.2 — Extended neurons (Izhikevich, ALIF, Synaptic, Alpha), EEG encoder, delta encoding
• v0.3 — NIR interoperability (export/import)
• v0.4 — Recurrent neurons (RLeaky, RSynaptic), learnable thresholds, expanded surrogates and losses
• v0.5 — Liquid State Machine, reservoir topology, mx.compile optimization
• v1.0 — Full documentation, comprehensive benchmarks, neuromorphic dataset loaders

Publications

• mlx-snn v0.1: Spiking Neural Networks on Apple Silicon via MLX (arXiv, 2026)
• mlx-snn v0.4: Spiking Neural Network Training on Apple Silicon: Cross-Framework Benchmarking (in preparation)

Citation

If you use mlx-snn in your research, please cite:

@misc{qin2026mlxsnn,
  title         = {mlx-snn: Spiking Neural Networks on Apple Silicon via {MLX}},
  author        = {Jiahao Qin},
  year          = {2026},
  eprint        = {2603.03529},
  archivePrefix = {arXiv},
  primaryClass  = {cs.LG},
  url           = {https://arxiv.org/abs/2603.03529}
}

Contributing

Contributions are welcome! Please open an issue or pull request on GitHub.

License

GPL-3.0