Fabio Guzman

Full Transformer into a custom chip. microGPT in RTL, generating names on a Virtex-5 FPGA at ~56k tokens/second.

CLIP-Finder enables semantic offline searches of images from gallery photos using natural language descriptions or the camera. Built on Apple's MobileCLIP-S0 architecture, it ensures optimal performance and accurate media retrieval.

Tool for visual profiling Core ML models, compatible with both package and compiled versions, including reasons for unsupported operations on the Neural Engine

Interpretability: Methods for Identification and Retrieval of Concepts in CNN Networks

Deep dive into Apple's AFMTextV7: 3.18B parameter foundation model with hybrid transformer architecture, native LoRA integration, and 1GB Apple Silicon footprint

How to create an OS from scratch

(forked) 🤓 Build your own (insert technology here) - added issue and website. PR welcome

Benchmark of Apple MLX operations on all Apple Silicon chips (GPU, CPU) + MPS and CUDA.

ESP32 helloserver

Scira (Formerly MiniPerplx) is a minimalistic AI-powered search engine that helps you find information on the internet. Powered by Vercel AI SDK! Search with models like Grok 2.0.

Writing & Making Operating System and Kernel parts so simple like Hello World Programs, Starting from writing Bootloaders, Hello World Kernel, GDT, IDT, Terminal, Keyboard/Mouse, Memory Manager, HDD ATA R/W, VGA/VESA Graphics

Small x86/ARM Operating System built from scratch

Trabajo Final de Sistemas Operativos: implementación de scheduler round-robin y miniShell interactivo (miniOS)

8 bit RISC Processor for SKY 130nm process

No description.

LaTeX templates for writing theses at Universidad Pontificia Bolivariana

Locally run an Instruction-Tuned Chat-Style LLM

Swift Core ML Examples

Instruct-tune LLaMA on consumer hardware

Go ahead and axolotl questions

Find out why your CoreML model isn't running on the Neural Engine!

ESP32 EEPROM routines for platformio

Esp32 webserver in Pplatformio

Call Traffic Analytics and Business Intelligence Platform.

On-device AI across mobile, embedded and edge for PyTorch

A Swift library that uses the Accelerate framework to provide high-performance functions for matrix math, digital signal processing, and image manipulation.

Reference implementation of the Transformer architecture optimized for Apple Neural Engine (ANE)