Matrix Multiplication with Half the Multiplications
Algebraic enhancements for GEMM & AI accelerators. Contribute to trevorpogue/algebraic-nnhw development by creating an account on GitHub.
Matrix Multiplication with Half the Multiplications launched on March 15, 2024 with 310 votes, #112 of 4,270 launches that month and more than 95% of that year's launches. The site was still online in September 2026, 2 years on.
In plain words
This is a software tool that optimizes matrix multiplication by reducing the number of computational multiplications required. It targets developers and researchers who work with matrix operations in mathematics, machine learning, and scientific computing. The approach cuts down multiplication operations while maintaining accuracy, making it useful for improving performance in computationally intensive applications. The code is available on GitHub for integration into existing projects.
written from the facts on this page · September 2026
Does a similar job
all alternatives →
Quadratic – Open-Source Spreadsheet with Python, AI (WASM and WebGL)2023 · quadratichq.com · ▲224Hi, I am David Kircos. The Founder of Quadratic (https://QuadraticHQ.com), an open-source spreadsheet application that supports Python, SQL (coming soon), AI Prompts, and classic Formulas. Unlike other spreadsheets, Quadratic has an infinite canvas (like Figma). As a result, you can pinch and zoom to navigate large data sets, and everything renders smoothly at 60fps. Our vision is to build a place where your team can collaborate on data analysis. You can write Python, AI Prompts, and Formulas in one spreadsheet feeding each other data and updating automatically. Quadratic is built…
Stella Nera – Maddness Hardware Accelerator2023 · github.com · ▲154Hashed Lookup Table based Matrix Multiplication (halutmatmul) - Stella Nera accelerator - joennlae/halutmatmul
Speeding up LLM inference 2x times (possibly)2024 · asciinema.org · ▲419Here's a project I've been working on for the last few months. It's a new (I think) algorithm, that allows to adjust smoothly - and in real time - how many calculations you'd like to do during inference of an LLM model. It seems that it's possible to do just 20-25% of weight multiplications instead of all of them, and still get good inference results. I implemented it to run on M1/M2/M3 GPU. The mmul approximation itself can be pushed to run 2x fast before the quality of output collapses. The inference speed is just a bit faster than Llama.cpp's, because the rest of implementation…
Boldly go where Gradient Descent has never gone before with DiscoGrad2024 · github.com · ▲232Trying to do gradient descent using automatic differentiation over branchy programs? Or to combine them with neural networks for end-to-end training? Then this might be interesting to you. We develped DiscoGrad, a tool for automatic differentiation through C++ programs involving input-dependent control flow (e.g., "if (f(x) < c) { ... }", differentiating wrt. x) and randomness. Our initial motivation was to enable the use of gradient descent with simulations, which often rely heavily on such discrete branching. The latter makes plain autodiff mostly useless, since it can only account for the…
More ai this month
the category →
Paintings by AI models: programs run through a simulation of oil paint.
AI · 20d ago · stillwet.art
PixVerse R2▲381A real-time world model you can explore and change
AI · 29d ago · world.pixverse.video
Source: Show HN post ↗
Launched alongside, March 2024
the whole month →
🌐 The Internet Computer! Free, Open-Source, and Self-Hostable. - HeyPuter/puter
Dev tools · 2024 · github.com
Microlaunch▲1,115Launch and get feedback on both the idea and product
Dev tools · 2024 · microlaunch.net
