Fine-grained parallelism with sub-nanosecond overhead in Zig

A conversational rabbit hole!

A massively parallel, high-level programming language

BlackHole is a modern macOS audio loopback driver that allows applications to pass audio to other applications with zero additional latency.

An implementation of bucketMul LLM inference

LLM training in simple, raw C/CUDA

Grok open release

A macOS application to run local models like Mistral, with the option to enhance conversations by switching to GPT-4.

Official Code for Stable Cascade

A Zig and Swift PoC for lightweight, resource-efficient native macOS apps.

All-in-one desktop app for running LLMs locally.

DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models

A simple vector store written in TypeScript with a WASM backend.

Letta (formerly MemGPT) is a framework for creating LLM services with memory.

A fast inference library for running LLMs locally on modern consumer-class GPUs

The TinyLlama project is an open endeavor to pretrain a 1.1B Llama model on 3 trillion tokens.

Visualizer for neural network, deep learning and machine learning models

Inference Llama 2 in pure Zig

Observable API proposal

Inference Llama 2 in one file of pure C

Take 3D stereoscopic screenshots in the visionOS emulator.

A reactive state machine implementation for OpenAI Chat Completions API.

CSS trick/bug to display a brighter white by exploiting browsers' HDR capability and Apple's EDR system

display a very bright white color on HDR-enabled displays

Run any open-source LLMs, such as Llama, Mistral, as OpenAI compatible API endpoint in the cloud.

An esbuild plugin that generates an HTML file.

Chat with GPT-4 and GPT-3.5 Turbo.

TypeScript for Tiny IoT Devices (ESP32, RP2040, ...)