An end-to-end AI platform that simplifies the orchestration of unstructured data, AI pipelines, and model deployment to build RAG and AI-first applications.

A multithreaded web crawler that converts website pages into markdown files to simplify document parsing and chunking for RAG and LLM fine-tuning.

A Dart MCP server and Flutter package that allows AI agents to inspect, drive, and interact with running Flutter apps in debug mode.

A Model Context Protocol server that gives AI agents access to Mapbox's location intelligence, enabling geospatial reasoning, routing, and interactive map visualization.

A library for sequence-to-sequence numeric prediction that allows predicting floating-point values from tokenizable inputs like text or images.

A unified deep learning framework based on PyTorch for ultra-large-scale sparse and dense computing, specifically optimized for industrial recommendation, advertising, and search systems.

A Rust client for the Hugging Face Hub API that allows developers to programmatically manage models, datasets, and spaces.

A kernel library for large head-dimension attention that optimizes SRAM and register usage to provide 1.5x-15x speedups over PyTorch SDPA for head dimensions up to 1024.

A Python library for flexible multimodal data preparation that optimizes the loading and manipulation of speech, audio, video, and image data for ML training.

An executable, Git-native catalog for tabular data work that turns ephemeral agent-generated scripts into reproducible, content-addressed pipelines.

A production-grade web scraping engine that provides clean markdown and anti-bot bypass for AI agents and LLMs.

A framework that uses pretrained Transformer language models to generate high-quality synthetic tabular data, supporting both training-based and schema-based generation.

A family of large-scale time series foundation models, including univariate and multivariate versions, for efficient and accurate forecasting.

A library built on Apache Spark for defining unit tests for large datasets to ensure data quality before it is used in applications or ML models.

A lightweight Python binding for LLVM that simplifies the creation of JIT compilers by providing a stable interface to LLVM's IR builder, optimizer, and JIT compiler APIs.

A pure Python, React-style framework for building scalable web applications and dashboards that work both in Jupyter Notebooks and as standalone apps.

A distribution and versioning system for AI assets like skills, prompts, and agents, allowing teams to share them across different AI tools using shared folders or Git vaults.

A codebase auditing skill for AI coding agents that provides structured, evidence-based reports across 25 dimensions including security, stability, and AI safety.

A tool to detect fake citations in academic papers by extracting references from PDFs and validating them against multiple academic databases.

A real-time security layer for AI agents that blocks destructive commands, scans plugins for malicious code, and performs daily security patrols to prevent data exfiltration and credential theft.

A fast, scalable machine learning system focused on online learning and reinforcement learning that uses the hashing trick to maintain a bounded memory footprint.

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

A deep learning atomistic model for simulating bulk materials across various elements, temperatures, and pressures to predict energy, forces, and stress.

TorchNPU is a PyTorch adapter plugin that enables the PyTorch framework to run on Huawei Ascend AI processors, providing optimized compute and distributed training capabilities.

Tilus is a tile-level GPU kernel programming language that enables high-performance, low-precision computation with explicit control over shared memory and registers.

A physics-informed machine learning tool for rapid airfoil aerodynamics analysis that is significantly faster and more robust than XFoil for design optimization.

A collection of AI inference runtimes for VapourSynth ML filters that enables optimized AI video processing across Intel, NVIDIA, AMD, and Apple hardware.

A bridge that lets you control official AI coding CLIs (Claude Code, Gemini, Grok, etc.) via Telegram, Matrix, and Slack without using API proxies.

A Python SDK for the Langfuse platform that provides observability, tracing, prompt management, and evaluation tools for LLM applications.

A multi-tenant vector and full-text search engine that uses object storage as its primary backend with a tiered RAM/NVMe cache for low-latency retrieval.

A Python library for LLM hallucination detection that uses various uncertainty quantification techniques to assign confidence scores to model outputs.

A unified Python framework for Scientific Machine Learning that simplifies the creation of PINNs, Neural Operators, and data-driven models.

A file-backed workflow and runtime for AI agents that replaces chat memory with repository-local artifacts to ensure continuity and consistency across sessions.

A collection of Model Context Protocol (MCP) servers and tools that enable AI agents to securely interact with IBM products and enterprise resources.

An interactive visualization tool for PyTorch models that allows users to trace and visualize the forward pass directly in notebooks.

codedb is a Zig‑based code‑intelligence engine that builds a local index of a repository (symbols, callers, dependencies, trigram search, etc.) and serves that information to AI coding assistants via the Model Context Protocol (JSON‑RPC over stdio). It provides MCP tools such as `codedb_context`, `codedb_explain`, and `codedb_callpath` for fast, sub‑millisecond queries, while remaining a read‑only “compass” for agents. Installable as native binaries or via an npm wrapper, it works on macOS, Linux and Windows, supports hybrid local/remote semantic retrieval, and can run as an HTTP server for debugging.

A formally verified programming language and secure runtime designed for machine-generated code, featuring an unambiguous syntax and a verified bytecode VM.

A security scanning tool for Model Context Protocol (MCP) servers and tools that uses YARA, LLMs, and Cisco AI Defense to detect malicious behavior and vulnerabilities.

A PyTorch library for training machine learning models with differential privacy, allowing users to protect sensitive training data while tracking privacy loss.

An engine-agnostic Java framework for deep learning that allows Java developers to build, train, and deploy models using a native Java experience.

An open-source agentic analytics platform that connects LLMs to enterprise data and tools, providing a governed environment for building, evaluating, and deploying data agents.

A resilient, HTTP-based data streaming protocol designed for AI agents and collaborative apps to ensure ordered, replayable, and resumable data delivery across devices.

A zero-dependency tool that pre-compiles codebase structure and documentation into compact context maps and wikis to drastically reduce token consumption for AI coding assistants.

A suite of development tools for streamlining the end-to-end cycle of LLM-based AI applications, from prototyping and prompt engineering to evaluation and production deployment.

An Electron-based desktop wrapper for DeepSeek Harness that provides a native window, built-in plugin market, and an interactive desktop pet.

A collection of system prompts and documentation from Xcode 27 beta, detailing Apple's internal instructions and strategies for its AI-assisted coding tools.

A Python package for handling annotated data matrices in memory and on disk, providing efficient support for sparse data and lazy operations.

A Python library for reading and processing data for training and evaluating JAX models, providing a fast and deterministic data pipeline.