A toolkit for exporting, authoring, and running open-source AI models on Apple silicon using the Core AI framework for macOS and iOS.

A high-performance TopK kernel implementation that provides 2x to 20x speedup over torch.topk for DeepSeek Sparse Attention and LLM sampling.

NumPy is the fundamental package for scientific computing with Python, providing N-dimensional array objects and tools for linear algebra and Fourier transforms.

A one-stop AI quantitative trading platform that provides tools and strategies for everything from learning and simulation to real-market deployment across multiple asset classes.

A library for efficient similarity search and clustering of dense vectors, capable of handling billion-scale datasets using CPU and GPU acceleration.

FlashMLA is DeepSeek’s CUDA library of dense and token‑level sparse attention kernels (including FP8 KV‑cache support) optimized for NVIDIA SM90/SM100 GPUs. It powers DeepSeek‑V3/V3.2/V4 models, delivering up to 660 TFLOPs on H800 GPUs and offering a fused kernel that combines Q‑norm, RoPE, attention and FP8 casting. Installation is via a standard pip build, and usage follows simple Python APIs (`flash_mla_with_kvcache`, `flash_mla_sparse_fwd`, etc.).

A deep learning optimization library that enables the training of models with trillions of parameters by optimizing memory and compute efficiency across distributed systems.

Chroma is an open-source vector database for AI that simplifies the storage, retrieval, and indexing of embeddings for similarity search.

A framework that bundles LLMs and speech-to-text tools into single-file executables for zero-installation, cross-platform local execution.

An open-source visual-first code editor for Next.js and TailwindCSS that allows users to design UIs visually in the browser and sync changes directly to the codebase using AI.

An open-source library for ingesting and pre-processing unstructured data from over 60 file types into structured formats optimized for LLMs and vector databases.

A scalable platform for running AI agent environments using Firecracker microVMs, featuring rapid snapshot-based booting and high-density resource management.

A real-time monitoring dashboard for NVIDIA DGX Spark and NVIDIA GPU Linux hosts that tracks hardware metrics and LLM inference performance.

An MCP server for searching and downloading academic papers from multiple open-access and public sources, designed for LLM-enhanced research workflows.

A utility to align session files and SQLite indices in Codex after switching AI providers, ensuring old chat sessions remain usable.

A persistent local memory server for AI agents using the Model Context Protocol (MCP), providing shared context, code graphs, and concept graphs across sessions.

ParadeDB is a Postgres extension that integrates full-text and vector search directly into the database, eliminating the need for a separate search engine.

Nx is a smart monorepo tool for TypeScript and polyglot codebases that uses incremental builds and AI-powered CI self-healing to accelerate development cycles.

A library providing state-of-the-art parameter-efficient fine-tuning methods to adapt large pretrained models using minimal compute and storage.

LiteParse is a fast, local, open-source PDF parsing tool that extracts spatial text and structure into Markdown or JSON without cloud dependencies.

An open-source, enterprise-ready embedded BI platform that combines interactive dashboards, paginated reports, and AI-driven natural language analytics.

A benchmark for evaluating large language models on their ability to resolve real-world GitHub software issues by generating functional code patches.

A governance framework for autonomous AI agents that provides deterministic policy enforcement, identity management, and audit logging to prevent unauthorized actions.

A hardware plugin that allows the vLLM inference engine to run on Huawei Ascend NPUs, supporting a variety of open-source LLMs and multimodal models.

A benchmark for evaluating how AI agents leverage and compose modular skills—instructions, scripts, and resources—to perform specialized workflows.

A multi-platform package manager and inference interface for Stable Diffusion that provides one-click installation and shared model management for various Web UIs.

A control plane and system of record for internal AI components that provides a centralized registry for discovering and deploying agents across multiple AI coding harnesses.

Hatchet is an open‑source orchestration engine that lets you run reliable background tasks, AI‑agent pipelines and durable workflows. It stores all state in PostgreSQL, provides retries, scheduling, DAGs, rate‑limiting, priority queues and a real‑time UI, and offers SDKs for Python, TypeScript, Go and Ruby. You can use it as a managed cloud service or self‑host it via Docker.

A Python package and command-line tool for web crawling and text extraction that turns raw HTML into structured data by removing noise and boilerplate.

A TypeScript framework for building and shipping MCP servers and interactive AI apps with native React views and built-in debugging tools.

A high-performance copy-paste detector for source code that identifies duplicated blocks across 224+ languages to help developers and AI agents refactor code.

An open-source data movement platform that provides 600+ connectors to sync data from APIs and databases into warehouses, lakes, or directly into AI agents as real-time tools.

A MySQL-compatible SQL database with built-in Git-like version control, allowing users to branch, merge, and commit changes to tables.

An MCP server that connects AI assistants to Devnors Data APIs, enabling them to search legal, enterprise, and academic data directly.

A Qemu-based emulator that boots iOS and macOS to a barebones root shell for kernel debugging and low-level Darwin system analysis.

A production-ready orchestration layer for running high-performance LLMs on-device across mobile, web, and IoT platforms using LiteRT.

A native desktop installer and manager for AI agent personas that synchronizes specialist personas across various AI coding tools like Cursor and Claude Code.

An MCP bridge that enables AI agents to control Adobe Premiere Pro, providing over 280 tools for automated video editing, project management, and timeline manipulation.

A multi-chain meme coin radar that scans for potential candidates using GMGN, GoPlus, and DexScreener to filter out high-risk tokens and provide evidence for manual verification.

A high-performance inference engine that allows developers to deploy AI models locally in their apps to eliminate latency, costs, and privacy concerns.

A production inference stack that decouples AI workloads from proprietary hardware, allowing a single codebase to run models on NVIDIA, AMD, Intel, and TPU/Trainium accelerators.

Apache Ossie is a vendor-agnostic semantic model specification that standardizes data definitions across AI agents, BI platforms, and data analytics tools to eliminate inconsistencies.

A desktop application for agent builders to prototype, trace, debug, and evaluate AI agents with local-first data management.

A native iOS companion app for the herdr agent-first terminal runtime that allows users to monitor and steer coding agents via a mobile agent console.

A library that ensures LLMs return reliable, structured JSON by using Pydantic for validation, type safety, and automatic retries.

An agentic skills framework for Godot 4.x that provides AI coding agents with domain-specific expertise and best practices for GDScript and C# game development.

A set of reliable PyTorch implementations of reinforcement learning algorithms designed to provide stable baselines for research and industry.

An open-source GPU cluster manager that orchestrates inference engines and provisions GPU instances for scalable AI model serving.