Explore the directory

A header-only C++ wrapper around GGML that simplifies loading GGUF or safetensors models, building computation graphs, and running them on CPU or GPU.

• Inference
• Libraries

An experimental retrieval-augmented generation system that ingests any MediaWiki site through its API and answers questions about that content through an OpenAI-compatible API, built as a Moodle Research project.

• RAG

An open source AI memory layer that scores how important a fact is and decays it over time at different rates, so a critical detail such as a medical allergy outranks a passing preference.

• Agent memory

A small language-model inference engine for consumer GPUs that implements continuous batching, prefix caching and per-request KV-cache control, invoked directly from Python in a few lines of code.

• Inference

A command-line tool for querying, writing and syncing a Tana workspace, including semantic search, delta sync through Tana's own Local API, and an MCP server for AI tool integration.

• MCP servers
• CLI tools

A lightweight, general-purpose harness for tool-using language model agents, meant as a fair execution substrate for agent benchmarks, a reference baseline, and a personal assistant runtime.

• Agent frameworks
• AI evals

A terminal AI agent, built on Elixir/OTP with a Rust interface, that classifies a request, filters out noise, and routes work to the right model, supporting any provider including an existing subscription.

• Coding agents
• LLM gateways

A PyTorch implementation of a published method for image classification using multi-agent reinforcement learning, where several agents move across an image and their policy is trained to classify it.

• Models
• Papers

A deprecated command-line configurator that detected a project's technology stack and generated matching configuration files for six different AI coding assistants from one interactive wizard.

• Coding agents
• CLI tools

A continuously reverified, machine-readable dataset of free-tier and trial-credit language model and adjacent AI APIs, with every entry dated, sourced to the provider's own documentation.

• Curated lists

A personal collection of more than 20 hands-on generative AI projects covering chatbots, retrieval-augmented generation, model fine-tuning and multi-agent systems, built while following a course.

• Tutorials

An ICML 2026 benchmark that tests whether AI code agents can handle data-intensive analytical tasks, requiring an agent to discover the right files among hundreds in a Linux sandbox before writing code.

• AI evals

A 52-file practical guide to AI covering running a first local model, retrieval-augmented generation, prompt engineering and building AI agents, organized from beginner to advanced.

• Tutorials
• Books

Open source file-transfer infrastructure that an AI agent can join on its own, getting its own email address, folder, inbox and API key so it can send and receive files without shared cloud credentials.

• APIs
• Backend

A hands-on curriculum of notebooks for building autonomous, stateful, goal-driven multi-agent systems on LangGraph, progressing from foundational concepts to multi-agent orchestration.

• Tutorials

A desktop app, CLI and server for running and coordinating many Claude Code sessions across several projects at once, breaking work into tasks, assigning them to sessions, and recording sessions for replay.

• Coding agents

An ontology-driven skill for the WorkBuddy agent that turns a business idea or requirements document into a runnable application, through staged requirements discovery, eleven domain-model types, and a generated SQLite-backed REST API, forms and knowledge graph.

• Agent skills

A drop-in vLLM plugin implementing Google's TurboQuant key-value cache quantization method, compressing the KV cache with asymmetric key/value bit-widths and validated across eight language models.

• Inference

Native Windows port of the NInfer C++/CUDA inference engine, built for a single RTX 5090 GPU with no WSL2 or Docker, serving text, image and video prompts through a local CLI or OpenAI/Anthropic-compatible HTTP APIs.

• Inference

A Python architecture and scaffolding tool that mines other projects' closed GitHub issues for recurring failure patterns before generating a new project's scaffold, then keeps auditing the growing codebase for dependency cycles, drift and fragile modules.

• CLI tools
• Code quality

A reproducible deployment bundle for running a distilled Gemma-based language model on a Xilinx Kria KV260 FPGA board, including the Verilog hardware sources, bitstream, tokenizer and integrity checks needed for edge inference.

• Inference

A desktop-native offensive security platform, built on Rust and Tauri, that combines Burp Suite-class web testing tooling with an autonomous AI agent driving the tools through a native MCP interface.

• Security

A collection of official examples and guides for building and deploying generative AI applications and agents on IBM's watsonx platform, covering several common agent frameworks.

• Agent frameworks
• Tutorials

An MCP server that connects Claude Desktop or Cursor to Chinese A-share and Hong Kong stock market data, exposing more than 30 financial data endpoints through natural language.

• MCP servers