Improved file parsing for LLM’s

A modular graph-based Retrieval-Augmented Generation (RAG) system

A tool to find sequencing data and metadata from public databases.

Convert PDF to markdown quickly with high accuracy

An elegant CLI wizard enhancing biotech research efficiency, with adaptable features for other domains, albeit with minor constraints.

LLM101n: Let's build a Storyteller

DSPy: The framework for programming—not prompting—foundation models

Simplifying reinforcement learning for complex game environments

SRA Tools

Examples in the MLX framework

Beautifully designed components that you can copy and paste into your apps. Accessible. Customizable. Open Source.

Video+code lecture on building nanoGPT from scratch

A markdown parser and compiler. Built for speed.

A nanoGPT pipeline packed in a spreadsheet

The Z3 Theorem Prover

A semantic-release plugin that updates package.json with the new release version.

Odysseus is a web scraping library built on top of Playwright, designed to handle dynamic web pages and CAPTCHA challenges with ease.

A minimal TypeScript library for fetching and parsing Google Scholar pages.

Inference code for Llama models

SkyPilot: Run LLMs, AI, and Batch jobs on any cloud. Get maximum savings, highest GPU availability, and managed execution—all with a simple interface.

A massively parallel, optimal functional runtime in Rust

A massively parallel, high-level programming language

Build AI Assistants with memory, knowledge and tools.

The official Meta Llama 3 GitHub site

LLM training in simple, raw C/CUDA

SWE-agent takes a GitHub issue and tries to automatically fix it, using GPT-4, or your LM of choice. It solves 12.47% of bugs in the SWE-bench evaluation set and takes just 1 minute to run.

🐚 OpenDevin: Code Less, Make More

OpenGFW is a flexible, easy-to-use, open source implementation of GFW (Great Firewall of China) on Linux

PyTorch code and models for V-JEPA self-supervised learning from video.