Semantic search over videos using Gemini Embedding 2 or Qwen3-VL.

A deepagent to witnessing slowdowns in your test runs.

Ponyfill of experimental React concurrent stores

Elegant HTTP Networking in Swift

JavaScript/TypeScript-native, low-boilerplate, object-capability RPC system

Framework to build resilient language agents as graphs.

A modern version of Monaco Editor.

Readymade evaluators for your LLM apps

Low output latency streaming HTML parser/rewriter with CSS selector-based API

Merge Python and JS code into one file that can be run in both languages.

Staging repo for development of native port of TypeScript

"Superposition" between sync and async

A WebAssembly runtime implemented in TypeScript types

Open-source infrastructure for Computer-Use Agents. Sandboxes, SDKs, and benchmarks to train and evaluate AI agents that can control full desktops (macOS, Linux, Windows).

A lightweight React Hook intended mainly for AI chat applications, for smoothly sticking to bottom of messages

A UI library to animate and style streaming LLM output

SDK for implementing Langgraph in your frontend in few seconds

Comparing WebSocket, WebRTC, and WebTransport under packet loss

A virtual monitor for screen sharing

Composable building blocks to build LLM Apps

Utilities for working with React Suspense

Things you can do with the token embeddings of an LLM

This repository showcases various advanced techniques for Retrieval-Augmented Generation (RAG) systems. Each technique has a detailed notebook tutorial.

container image to single executable compiler

LLM101n: Let's build a Storyteller

Turn any PDF or image document into structured data for your AI. A powerful, lightweight OCR toolkit that bridges the gap between images/PDFs and LLMs. Supports 100+ languages.

Understand React Server Components and Server Actions by building a framework with them.

A simple React Server Components implementation that you can build yourself 🙌

⛩️ The minimal React framework