An MCP server that enables AI models to convert documents between multiple formats using Pandoc, supporting everything from Markdown to PDF and DOCX.

A local-first runtime for AI agents that provides sandboxed code execution, persistent session management, and tool governance on the user's own infrastructure.

A knowledge base and plugin for AI coding tools that provides verified reference guides and source code to help agents generate accurate, high-performance Jetpack Compose code.

A Ruby implementation of the Model Context Protocol (MCP) server that allows LLMs to analyze and interact with Rails projects through structured tools and documentation.

A full-stack starter template for AgentCore that provides a secured React frontend and AWS infrastructure to accelerate the deployment of AI agent applications.

A high-performance library for running Generative AI models locally on PCs and laptops, providing optimized pipelines for text, image, and speech models on Intel hardware.

A curated registry of Model Context Protocol (MCP) servers that allows developers to discover and deploy secure, containerized tools for AI assistants.

A Python library for generating scalable synthetic data within Databricks using Spark, designed for testing, benchmarking, and demos.

A Kubernetes operator that enables AI-driven cluster analysis and troubleshooting, featuring automated remediation for specific workload failures.

A Swift library for integrating various AI APIs into apps with an optional secure proxy backend to protect API keys and manage rate limits.

An MCP server that gives LLMs read and write access to Airtable databases, enabling them to manage records, schemas, and comments.

A scalable asynchronous reinforcement learning framework for LLM agents that uses in-flight weight updates to maximize GPU utilization while maintaining on-policy data freshness.

A lightweight, standalone C++ inference engine for Google's Gemma models, designed for research and experimentation with a minimalist, CPU-optimized implementation.

A high-performance serving stack for DeepSeek-V4-Flash on NVIDIA DGX Spark, featuring continuous batching, speculative decoding, and a memory governor for massive context windows.

A cache management framework for LLMs that reduces inference latency and GPU memory usage by persisting KV caches and implementing sparse attention retrieval.

A Vue-native framework for building interactive terminal user interfaces (TUIs) using components, Flexbox layout, and reactive state.

A framework for RAG pipelines that automatically selects the best chunking strategy for each document by evaluating it against five intrinsic quality metrics.

A centralized collection of LangChain integrations for Google's Generative AI, Gemini API, and Vertex AI services.

An Emacs integration for Aider, bringing AI pair programming and automated code editing to the Emacs editor.

A production-grade deployment toolkit for LLMs and VLMs based on PaddlePaddle, offering high-performance inference and broad hardware compatibility.

A Python library for PyTorch that provides tools for the efficient loading, preprocessing, and augmentation of 3D medical images.

A neural network library for JAX that enables object-oriented model definition while preserving JAX's pure function transformations.

A pure-Julia machine learning framework that provides lightweight abstractions for building and training models using native GPU and automatic differentiation support.

Cube Studio is an integrated development environment for data science and AI, focusing on providing consistent environments across the development lifecycle.

A high-performance, lossless HTML-to-Markdown converter with Rust core and bindings for 16 languages, designed to handle malformed real-world HTML.

LlamaDeploy is a tool for deploying and serving AI workflows, though it is now deprecated in favor of llama-agents.

Acme is a research framework for reinforcement learning that provides scalable building blocks and baseline agents for developing and testing RL algorithms.

A high-performance Mixture-of-Experts (MoE) implementation optimized for NVIDIA Hopper and Blackwell GPUs to reduce memory usage and increase training throughput.

A minimal PyTorch re-implementation of Qwen 3.8 that provides a readable codebase and int4 quantization support for easier model deployment.

stable-pretraining is a PyTorch Lightning-based framework for foundation-model research that simplifies self-supervised, supervised, and multi-modal pretraining (SimCLR, DINO, MAE, CLIP, etc.) via a dict-based design, live evaluation callbacks, GPU-side augmentation, and SLURM-grade orchestration, with an experimental JAX backend.

A C++ library providing unified wrappers for SIMD intrinsics to accelerate numerical computations across different CPU architectures.

A Linux kernel driver for the Intel® NPU, enabling energy-efficient AI inference acceleration on Intel® Core™ Ultra CPUs.

A hardware plugin that enables vLLM to run on Kunlun XPU, providing optimized kernels and support for a wide range of LLMs and multimodal models.

A lightweight multimodal LLM inference engine optimized for mobile and edge devices, supporting various NPUs and GPUs with advanced quantization.

A software library that combines tree-boosting with Gaussian processes and mixed-effects models to handle non-linearities and data dependencies like spatial or grouped effects.

A runtime security platform for AI applications that detects prompt injection, manages token budgets, and blocks bots and sensitive data leaks.

A GPU-optimized platform for building Large Population Models (LPMs) that simulate millions of interacting agents to model complex societal dynamics.

A W3C specification for the Web Neural Network API, providing a standardized way for web applications to access hardware acceleration for machine learning models.

A practical Python library implementing CMA-ES and its variants for black-box optimization of continuous, integer, and categorical variables.

A pure Python machine learning library built from scratch for educational purposes, providing transparent implementations of core ML algorithms and an autograd engine.

A semantic code search and symbol intelligence system that provides AI coding assistants with deep codebase context, memory, and navigation tools via MCP.

A high-performance LLM inference engine written natively in C#/.NET, providing native CPU/GPU support for transformer models without relying on Python or llama.cpp.

An MLOps/LLMOps scheduler and orchestration tool that automates the execution of ML experiments across local and cloud resources with zero configuration.

An automated machine learning framework that allows users to create ML models via ready-to-use presets or by building custom pipelines for tabular, text, image, and time series data.

An open-source intelligent operations (AIOps) Agent platform that integrates observability, event tracking, and task execution into a controlled, AI-driven workflow for IT operations.

A Go-based tool to parse GGUF model metadata and estimate memory usage and performance without requiring full file downloads.

A high-performance system for deploying multiple AI models on shared GPUs, featuring ultra-fast checkpoint loading and GPU multiplexing to enable efficient serverless inference.

A high-performance Llama 2 inference implementation written in pure Mojo, leveraging SIMD and vectorization to optimize CPU performance.