Explore the directory

A library that reduces the memory needed to run very large language models, letting models such as a 70-billion-parameter Llama or larger mixture-of-experts models run on a single small-VRAM GPU.

• Inference

An open-source LLM engineering platform for tracing, evaluating and debugging AI applications, with prompt version management and an interactive playground for iterating on prompts.

• AI evals

A self-hosted personal AI assistant that runs locally or in the cloud, reachable from many chat platforms at once, with layered memory, a plugin and skill marketplace, and sandboxed tool execution.

• Agent frameworks

A project template that has an AI coding agent recreate any website as a clean Next.js application, using a multi-phase pipeline of screenshots, extracted design specs and parallel builder agents.

• Agent skills
• Web frameworks

A DeepSeek-native coding agent for the terminal, packaged as a single Go binary, reachable from a CLI, a desktop app, a browser, or an editor over the Agent Client Protocol.

• Coding agents

A tool for iteratively optimizing, testing and comparing AI prompts across multiple models, available as a web app, desktop app, Chrome extension or Docker deployment.

• AI evals

A retrieval-augmented generation engine that replaces a vector index with a tree structure of a document, letting an LLM reason its way to the relevant section the way a person would.

• RAG

A coding agent available as a CLI, VS Code extension and JetBrains plugin, configurable to work with a chosen model and workflow.

• Coding agents

A Microsoft research project and data pipeline that extracts structured, graph-based context from unstructured text using LLMs, to help a language model reason over private data.

• RAG

A high-performance serving framework for large language models and multimodal models, aimed at low-latency, high-throughput inference across NVIDIA and AMD GPUs as well as TPUs.

• Inference

A self-hosted AI answering engine that searches the web and cites its sources, combining local models through Ollama or cloud providers with a bundled private web search backend.

• RAG

An agent memory system built to let agents learn from experience over time, rather than only recall past conversations, reporting strong results on a long-term conversational memory benchmark.

• Agent memory

An open-source Linux server management panel with a web interface for host monitoring, file, database and container management, plus a built-in AI gateway for managing agents and models.

• Hosting

A cross-platform framework for building customizable on-device machine learning pipelines for vision, audio and text, deployable to mobile, web, desktop and edge devices.

• Models
• Inference

Anthropic's official, curated marketplace of Claude Code plugins, listing internal plugins built by Anthropic alongside reviewed third-party plugins, installable directly from Claude Code.

• Agent skills
• Curated lists

A hardware-detection tool that inspects a machine's CPU, RAM, GPU and VRAM to recommend which open-source language models, and which quantization, will actually run well on it.

• CLI tools

A local gateway that gives coding agents such as Claude Code, Codex and others one stable endpoint, while letting a developer manage providers, models, routing rules and fallbacks from a single place.

• LLM gateways

A self-hostable personal AI application that chats with any local or hosted language model, answers questions from the web or a user's own documents, and can run scheduled research automations.

• RAG
• Inference

A free, self-hostable platform-as-a-service that deploys applications and databases as an open-source alternative to Vercel, Netlify and Heroku.

• Hosting

A Java diagnostic tool that lets developers troubleshoot production issues, such as classloader conflicts and slow method calls, without modifying code or restarting the JVM.

• Debugging

A frontend SDK for building agent-native applications: generative UI, shared state, and human-in-the-loop workflows for React, Angular, Vue and React Native, plus Slack and Microsoft Teams.

• Agent frameworks
• Web frameworks

A Model Context Protocol server that gives large language models browser automation through Playwright's accessibility tree, so agents can interact with web pages without screenshots or vision models.

• MCP servers
• Browser automation

A pure C inference engine that runs very large mixture-of-experts language models on consumer hardware by treating disk, RAM and VRAM as one memory tier.

• Inference

Directus wraps an existing SQL database with automatically generated REST and GraphQL APIs and a visual admin Studio, so engineers, non-technical teammates and AI agents can all work with the same live data.

• Backend
• Databases