Explore the directory

A curated survey repository of research, benchmarks and products on coding agents that improve themselves from execution outcomes, memory, tools and accumulated experience.

• Coding agents
• Curated lists

A curated collection of datasets for training and evaluating large language models, covering medical AI, NLP, multimodal learning, instruction tuning, reasoning, code generation and evaluation benchmarks.

• Models
• Curated lists

Hand-written CUDA kernels that run NVFP4 and FP8 quantized weights for the Qwen3.8 model family on older Tesla V100 GPUs, serving them through an OpenAI-compatible endpoint.

• Inference

A Model Context Protocol server that lets an LLM client query and manage a Kubernetes cluster, including custom resources and Helm releases.

• MCP servers
• Containers

A native desktop UI for Claude Code that adds orchestration, live interleaved thinking, background agents and multi-provider support, built with Tauri and React.

• Coding agents

A Python tool that fine-tunes Qwen3-TTS on a few minutes of audio to clone a voice and generate speech locally on your own GPU, from a CLI or web UI.

• Speech

An asynchronous Python client for the Model Context Protocol with an interactive CLI and a reactive web UI, connecting AI models to MCP servers over stdio and SSE.

• MCP servers

An Android app for running large language models locally on-device, with document Q&A, vision input, tool use and background model downloads.

• Inference
• Mobile

An open-source tutorial curriculum that teaches how to build AI agent systems from first principles, from foundational theory through production-grade multi-agent applications.

• Tutorials

An open-source Python framework for fine-tuning retrieval-augmented generation systems across both centralized and federated architectures.

• RAG
• Models

An AWS sample platform for autonomous background coding agents that turn a submitted task into a reviewed pull request, running each task in an isolated cloud environment with orchestration, observability and cost controls.

• Coding agents
• Hosting

A Python command-line tool that downloads models from the Ollama library without needing Ollama installed.

• Inference
• CLI tools

An MCP server that gives AI agents persistent memory: semantic search and recall over stored memories, confidence-scored verified facts, and encrypted sync between machines.

• MCP servers
• Agent memory

A native macOS desktop app that starts, stops and duplicates development projects with one click, with a built-in terminal for running AI coding agents such as Claude Code and Codex alongside project services.

• Terminals
• Desktop utilities

A local-first video editing MCP server, Python library and CLI that lets AI agents trim, caption and repurpose video with FFmpeg, backed by typed tools and quality checks instead of invented FFmpeg flags.

• MCP servers
• Generative media

A declarative, Rust-based manager for AI agent environments that syncs skills, MCP servers, commands and instructions across projects, machines and multiple coding agents from one YAML config.

• Agent skills

A Python tool that deploys Haystack pipelines and agents as REST APIs, MCP tools, or A2A services, with optional OpenAI-compatible chat endpoints and a bundled Chainlit UI.

• MCP servers
• RAG

An Android app that runs large language models on-device using Google's LiteRT-LM runtime, with GPU acceleration, cloud-provider integration, persistent memory and offline voice interaction.

• Inference
• Mobile

A typed Node.js CLI that lets an AI agent edit raw video footage conversationally, transcribing clips, reasoning over an edit-decision list, and rendering the result with ffmpeg.

• Generative media
• CLI tools

A local, intelligent LLM gateway that profiles installed Ollama and llama.cpp models on your own hardware and automatically routes each request to whichever model performs best for it, as a drop-in OpenAI-compatible endpoint.

• LLM gateways

A text-first command-line tool for managing and running MLX models on Apple Silicon, covering listing, cloning, converting, quantizing and serving, with curated support for select vision and audio models.

• Inference
• CLI tools

An MCP server that gives an AI agent a full graphical desktop it can see and control, moving the mouse, typing, launching apps and running shell commands, in a sandboxed virtual desktop or on the machine in front of you.

• MCP servers
• Containers

A reference implementation of a ternary-weight tiny language model with a bit-exact Python-to-C99 inference engine, sized to run inside a microcontroller's RAM budget.

• Models
• Inference

A cross-harness platform for building portable teams of AI agents, called Missions, that combine models, tools, shared context and memory once and run unchanged in Codex, Claude Code, Pragma's own desktop app, or an embedding SDK.

• Agent frameworks