anykrver

anykrver / neuraedge-

Public

A minimal neuromorphic chip design in SystemVerilog to learn how brain-inspired computing works from the ground up.

18
1
100% credibility
Found Mar 16, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
SystemVerilog
AI Summary

NeuraEdge is an open-source educational project implementing a simple brain-inspired processor to demonstrate spike-based event-driven computing for tasks like pattern recognition and digit classification.

How It Works

1
🔍 Discover NeuraEdge

You find this exciting educational project that teaches how chips can work like brains for super-efficient computing.

2
🛠️ Get your computer ready

Install a few free simple tools so you can run the brain chip simulations right on your machine.

3
🧪 Test a single brain cell

Run a quick simulation to watch one neuron fire spikes just like in a real brain.

4
🧠 Solve tricky puzzles

Try examples like figuring out XOR logic or recognizing simple patterns, seeing the chip learn and decide.

5
✍️ Read handwritten numbers

Train the chip on digit images and watch it classify numbers with brain-like spike patterns.

6
⚡ Try on real hardware

Optionally load it onto a small circuit board to feel the power of live brain-inspired processing.

🎉 Unlock brain computing secrets

You've built and understood a mini brain chip that runs tasks way more efficiently than regular computers.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 18 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is neuraedge-?

NeuraEdge is a minimal SystemVerilog design for a neuromorphic chip that implements brain-inspired computing from the ground up, using leaky integrate-and-fire neurons, spike routing, and STDP learning. It lets you simulate event-driven spiking neural networks for tasks like XOR classification, pattern recognition, and MNIST digit recognition, then synthesize to FPGAs like Basys 3. Developers get ready-to-run examples with Python training kernels that export weights directly for hardware.

Why is it gaining traction?

Unlike complex commercial neuromorphic chips or software-only SNN frameworks, NeuraEdge strips everything to core principles—no bloat, just a synthesizable baseline that fits on cheap FPGA boards and runs at 160MHz with under 4% resource use. The hook is its make-based simulation workflow with Icarus Verilog and GTKWave waveforms, plus Python scripts for training and validation, making brain-inspired hardware accessible without Vivado expertise.

Who should use this?

FPGA engineers dipping into edge AI neuromorphic designs, hardware researchers prototyping spike-based classifiers, or university students learning SystemVerilog for brain-inspired chips. Ideal for those evaluating minimal neuromorphic computing before scaling to Loihi-like systems.

Verdict

Grab it if you're curious about neuromorphic hardware fundamentals—docs are thorough, tests cover unit to full-chip, and MNIST hits ~90% accuracy post-training. With 18 stars and 1.0% credibility score, it's early-stage but rock-solid for learning; expect to tweak for production.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.