Explore the directory

A tool that lets you start, resume, approve, interrupt, and monitor a local Codex coding session from a Discord thread on your phone or browser.

• Coding agents

An openly available handbook covering reinforcement learning from multi-armed bandits through modern policy-gradient and actor-critic methods, with runnable code examples for each chapter.

• Tutorials
• Books

An open-source enterprise AI middleware platform for connecting APIs, databases and file storage to any LLM, and building authenticated, observable AI agents and workflows on top of the flo-ai framework.

• Agent frameworks
• Backend

A one-command, pinned serving stack that runs the Qwen3.8 model family on a single NVIDIA DGX Spark, with several switchable, benchmarked speed and quantization configurations.

• Inference

A C++ library that extends CppAD to generate C/C++ source code for computing derivatives of mathematical models, combining operator-overloading with compiled, source-generated derivatives.

• Libraries

An interactive 3D visualizer that renders a real transformer's forward pass, showing attention heads, RoPE rotations and the residual stream update live as a small model like Qwen2.5-0.5B or a local GGUF file runs.

• Tutorials

A free, open documentation platform built with Laravel and Filament, enhanced with optional private AI features powered by a self-hosted Ollama or OpenClaw model.

• Documentation tools

A terminal dashboard and swarm orchestrator for Claude Code sessions, with a small local-LLM brain that learns your preferences and auto-approves or denies tool calls.

• Coding agents
• Terminals

The companion code repository for a hands-on LLM engineering course, covering running local models with Ollama and building agentic AI projects.

• Tutorials

A performance-tuned build of llama.cpp for the AMD Strix Halo APU, bundling Flash Attention and MoE-prefill fixes with a current Mesa driver to speed up long-context, quantized-KV inference.

• Inference

A drop-in proactive memory harness for AI agents that builds a compaction tree over past sessions and combines it with keyword or vector search, so an agent can surface relevant past context that nobody explicitly asked about.

• Agent memory

A single Rust binary that tracks a project's specification as a flat list of atomic, taggable claims, some backed by a shell command, so an AI coding agent and a developer can verify what is actually built.

• CLI tools
• Testing

A framework for turning small language models into capable autonomous agents, with support for nine cloud providers, four local backends, 66 built-in tools and a run ledger that tracks cost and where time went.

• Agent frameworks

A KV-cache compression method for LLM inference that improves on an existing method by exploiting a structural property shared across models, described in a NeurIPS 2026 submission.

• Inference

A CUTLASS-based quantized BLAS library for low-bit deep learning inference and training on NVIDIA Blackwell GPUs, supporting MXFP4, MXFP8 and NVFP4 microscaling formats.

• Inference
• Libraries

An MCP server that indexes a documentation set by heading hierarchy and retrieves a single section, heading subtree or ranked search result, instead of loading whole files into an agent's context window.

• MCP servers
• Documentation tools

A curated list of papers, code, and resources for speech language models and end-to-end speech dialogue systems, covering understanding, benchmarks, and full-duplex modeling.

• Speech
• Papers

A multi-agent harness that autonomously tunes LLM and diffusion inference serving configurations, frameworks and GPU kernels for AMD Instinct GPUs, then stores proven results in a reusable recipe knowledge base.

• Agent frameworks
• Inference

A project building a knowledge base about the Windows Registry, together with a reusable Python module for working with Windows Registry resources.

• Libraries
• Documentation tools

A language server that brings IDE features for the V programming language, such as go-to-definition, code completion, and hover documentation, to VS Code, Vim, and other editors.

• Editors

A self-hosted knowledge workspace for teams that organizes documents, notes, and a knowledge graph for retrieval-augmented question answering, with built-in RAG evaluation and nightly self-consolidation.

• Agent memory
• RAG

A C++ inference engine built on llama.cpp for Vision-Language-Action robot policies, running open VLA models such as SmolVLA, GR00T and BitVLA as self-contained GGUF files without Python.

• Inference

An experimental storage-aware mixture-of-experts inference stack that prefetches predicted experts before decoding starts, aimed at keeping SSD-offloaded MoE models from stalling generation.

• Inference

A zero-dependency Python CLI that manages MCP tools and agent skills for every AI coding agent from one config, cutting the tokens spent on tool and skill discovery by keeping full descriptions on disk instead of in context.

• MCP servers
• CLI tools