Explore the directory

An open-source retrieval-augmented generation platform that separates document ingestion, multimodal indexing, hybrid retrieval and agent integration into independently deployable services, with tenant isolation, role-based access control and citation-ready results.

• RAG

Workshop materials teaching security practitioners how to write, refine and verify prompts for real security tasks, including a hands-on prompt-injection challenge and packaging prompts into reusable Modelfiles.

• Security
• Tutorials

Research code and results accompanying a paper showing that a transformer's KV cache can be edited in place or recombined across contexts instead of always being recomputed, with large throughput and latency gains.

• Papers

An experimental operating-system-styled runtime in which every component is an autonomous agent: natural language replaces syscalls, Hebbian-learning-based attention replaces the scheduler, and destructive operations require multi-agent consensus instead of file permissions.

• Agent frameworks

Course material for an O'Reilly live training on running open-weight decoder LLMs locally, with notebooks comparing the transformers library, vLLM, SGLang, tabbyAPI and llama.cpp as execution engines.

• Tutorials

A portable, self-contained package that bundles Ollama, Hollama and Caddy into one folder, giving a no-install web chat interface for running local LLMs on Windows or macOS.

• Inference

A self-hosted multi-agent framework that lets you command AI agents with different specialties for coding, data retrieval and analysis from Telegram, Discord or Slack, coordinated by a persistent grep-based memory system and a standardized agent harness.

• Agent frameworks

A document retrieval-augmented generation system that can run entirely client-side in the browser or as a FastAPI and FAISS backend, with fair-share retrieval across multiple uploaded documents and LaTeX math rendering.

• RAG

A small proxy that translates OpenAI-compatible chat requests into Mistral's native /v1/conversations endpoint, letting the 9router gateway reach a GLM-5.2 model through Mistral without hitting the rate limits on Mistral's OpenAI-compatible endpoint.

• LLM gateways

A command-line control plane that installs, authenticates, diagnoses and connects the XMemo memory service across editors, CLIs and autonomous AI agents through a hosted or local MCP server.

• Agent memory
• CLI tools

A self-hosted MCP server that gives AI agents and teams adaptive long-term memory beyond simple RAG, combining hybrid semantic and keyword search with a neural memory graph that strengthens connections between related memories as they are used.

• MCP servers
• Agent memory

A Windows desktop app that turns ideas into tasks, runs coding agents such as OpenCode, Claude Code, Codex, Grok and Antigravity CLIs against a chosen project folder, and tracks their work with a live session tree and a review queue.

• Coding agents

An example application pairing a LangGraph agent that connects to MCP servers and calls their tools with a CopilotKit frontend for UI and state synchronization.

• MCP servers

A self-hosted client that runs Claude Code, Codex, opencode and Kimi Code in tmux on your own Linux or Mac server, using your existing CLI subscriptions, and lets you drive and monitor them from a desktop app, phone app or browser.

• Coding agents

A stateless policy kernel that gates AI agent tool calls with five verdicts (allow, deny, dry run, approve required or transform), adds crash-safe checkpoints and a tamper-evident audit trail, and can wrap any agent harness or run a minimal loop of its own.

• Agent frameworks
• AI evals

A command-line local AI model manager for macOS and Linux that stores models once in the Hugging Face cache and exposes them to tools such as LM Studio, llama.cpp, Jan and ComfyUI through symlinks, avoiding duplicate downloads.

• Inference
• CLI tools

A .NET library for on-demand local AI inference that ships with no bundled models, downloading and caching embeddings, generation, vision, OCR, translation, transcription, speech-synthesis and image-generation models the first time they are used.

• Inference
• Libraries

A set of examples for deploying LLMs on Aalto University's Triton cluster, covering local inference through Hugging Face Transformers and vLLM, Aalto's own LLM API, and configuring AI coding agents against Aalto's endpoints.

• Tutorials

A pure-Python decision engine that scores LLM responses for attribution, specificity, relevance and context quality to detect hallucinations and output an accept, review or reject verdict.

• AI evals

A mechanistic-interpretability research project testing whether a language model's internal activations can reveal the operation and operands of an arithmetic prompt well enough to route it to a calculator without parsing the prompt text.

• Papers

A self-hosted, embeddable chatbot that indexes a website's own content and answers visitor questions over text or voice, built entirely on the Cloudflare Workers stack with Vectorize for retrieval.

• RAG

A byte-identical Rust port of kimi-k3-in-c that runs CPU-only inference for the 2.78-trillion-parameter Kimi K3 mixture-of-experts model, with no GPU, BLAS or PyTorch, in an 8 GB memory floor.

• Inference

An Android Flutter app and Termux CLI package that sets up a proot Ubuntu environment to run the OpenClaw AI gateway on a phone, with a built-in terminal, web dashboard and SSH access, and no root required.

• Agent frameworks

An MCP server that computes git repository analytics such as hotspots, churn, temporal coupling, knowledge maps and risk scoring entirely from local commit history, for AI coding agents like Claude Code and Codex.

• MCP servers
• Code quality