Explore the directory

Open-source infrastructure for world models, unifying inference and benchmark evaluation across video generation, 3D/4D representation and embodied-action models behind one TUI, CLI and Studio interface.

• AI evals
• Generative media

A cross-platform SDK for running compressed large language models fully on-device, using a quantization algorithm that learns an optimal per-weight bit allocation instead of a fixed scheme.

• Inference

An MCP server that gives AI assistants access to the Zerodha Kite Connect trading API, for retrieving market data, managing a portfolio and placing orders.

• MCP servers

An open-source memory runtime for AI agents that compiles raw events into typed, provenance-tagged memories and assembles deterministic, token-bounded context bundles instead of retrieving them fresh at query time.

• Agent memory

Course code for an O'Reilly live training and video series on building and operating AI agents, covering agent frameworks such as CrewAI, LangChain and AutoGen from prototype through production.

• Tutorials

A lightweight, single-binary coding agent harness written in Go, with built-in providers for dozens of LLM services, a terminal UI, and support for extensions, themes and portable agent files.

• Coding agents

A machine learning platform for Apple Silicon, written in Rust, covering local LLM inference, LoRA/QLoRA fine-tuning, quantization and serving through a desktop GUI, a terminal TUI and Metal/Neural Engine acceleration.

• Models
• Inference

A lossless compression utility that uses a user-provided LLM as the probabilistic model for an arithmetic coder, achieving high compression ratios on structured or natural-language text.

• CLI tools

A collection of Python and Node.js example code from the "Code AI with Rok" YouTube tutorial series, covering LLM providers, agent frameworks, cloud AI services and vector databases.

• Tutorials

A curated collection of code samples and tutorials for building AI agents on Google's Gemini Enterprise Agent Platform, formerly Vertex AI, covering its models, frameworks, protocols and hosting tools.

• Tutorials

A menu-bar screenshot and annotation app for macOS and Linux that also gives AI coding agents a visual mode, rendering diagrams and previews for a user to review and annotate on screen.

• Coding agents
• Desktop utilities

A local-LLM-first autonomous coding agent that turns a plan into acceptance criteria as an immutable contract, then implements against that contract, running on Ollama, vLLM, sglang or llama.cpp.

• Coding agents

An MCP server that connects Claude Desktop or Claude Code to a Monarch Money personal-finance account, giving an AI assistant access to accounts, transactions, budgets and analytics.

• MCP servers

An open-source, model-neutral coding agent with a desktop app and a terminal CLI/TUI, built in Rust, for connecting to OpenAI-compatible, Anthropic-compatible, DeepSeek, Qwen, Kimi or private model endpoints.

• Coding agents

A TypeScript sample application demonstrating the Retrieval Augmented Generation pattern on Azure, using Azure AI Search for retrieval and Azure OpenAI to power a ChatGPT-style Q&A experience over a company's own documents.

• RAG
• Tutorials

InstantCMS is a free, open-source content-management system for building websites of any type, combining flexible content types with built-in social-networking features such as user profiles and community tools.

• Backend

A Swift vector database for on-device apps, combining local vector storage and hybrid search with pluggable embedding providers, including Apple's MLX and NaturalLanguage frameworks.

• Databases

A graph and vector database built on Neo4j that accepts any JSON and returns typed, searchable, relationship-aware records without a schema or migrations, meant as a memory layer for AI agents.

• Agent memory
• Databases

An open-source MCP server that lets AI coding agents in different harnesses, such as Claude Code, Codex, Cursor and Gemini CLI, message each other, detect file-edit overlaps, and hand off tasks.

• Coding agents
• MCP servers

An MCP server, CLI and Codex skill that connects AI assistants to Korean retail chains, convenience stores, movie theaters, gas prices and nearby-place search, without needing separate API keys.

• MCP servers

An enterprise knowledge-base platform that combines retrieval-augmented generation with an AI agent, answering questions from uploaded documents with citations back to their sources.

• RAG

A free Taiwan legal MCP server and CLI that retrieves Taiwan court judgments, administrative interpretations and constitutional court rulings by semantic search, for use with the caller's own AI tool.

• MCP servers

An LLM-powered agent that automates Google Dorking for bug bounty hunting and penetration testing, using a search-engine API and a choice of LLM provider.

• Security

A lightweight C++ SDK implementing the Model Context Protocol, letting AI models and agents interact with external resources and tools over JSON-RPC 2.0.

• MCP servers