Analysis updated 2026-07-26
Simulate how a light-based CPU performs basic arithmetic like addition.
Generate laser instructions to physically carve optical circuits into a crystal.
Test a tiny operating system that manages memory and schedules processes for an optical chip.
Experiment with a neural network compiler designed for photonic hardware.
| weizhishu123/jelly-computer | 0xustaz/streamgate | a-bissell/unleash-lite | |
|---|---|---|---|
| Stars | 1 | 1 | 1 |
| Language | Python | Python | Python |
| Setup difficulty | easy | hard | hard |
| Complexity | 4/5 | 4/5 | 4/5 |
| Audience | researcher | developer | researcher |
Figures from each repo's GitHub metadata at analysis time.
Requires Python 3.7 and the NumPy library, with PyTorch as an optional dependency for neural network features.
This repository is an experimental computing project that replaces traditional electronic transistors with light-based optical circuits. Instead of electrons flowing through silicon chips, it uses laser pulses traveling through channels carved inside a transparent crystal called lithium niobate. The physical shape of those channels, specifically a Y-shaped branching pattern, encodes basic arithmetic operations like addition, subtraction, multiplication, and division. The repository contains a full vertical stack of software, from physics simulation up to a small operating system. At the bottom is a physics simulator that models how light behaves in these Y-shaped waveguides. Above that, a code generator produces instructions for a femtosecond laser, the machine that would physically carve the channels into the crystal. On top of that sit two CPU simulators, a 5-chip version and a 7-chip version, each with 17 instructions including arithmetic, memory operations, jumps, and interrupts. At the highest level, a micro operating system provides memory management, process scheduling, a file system, eight system calls, and an interactive shell. The project is currently a simulation and design toolkit. No physical jelly computer has been built yet. The README explicitly invites someone to manufacture the first physical optical gate using the generated laser instructions and demonstrate that 3 plus 5 equals 8. The first person to do so would be recognized as the builder of the first photonic jelly computer. To run the software, a user needs Python 3.7 or newer and the NumPy library. An optional dependency on PyTorch is mentioned for a neural network compiler. The project is released under the MIT license.
An open-source design toolkit for building a computer that runs on light instead of electricity. It includes simulators and a tiny operating system to model optical computing circuits before they are physically built.
Mainly Python. The stack also includes Python, NumPy, PyTorch.
Use this code freely for any purpose, including commercial projects, as long as you include the original copyright notice.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly researcher.
This repo across BitVibe Labs
Verify against the repo before relying on details.