A suite of skills and a plugin that offloads routine agent decisions (like model routing and result filtering) to a low-cost decision model to reduce latency and token spend.

A Codex skill that optimizes AI software development by using Astra for high-level planning and review, while delegating high-volume implementation to DeepSeek V4.1 Flash to reduce costs.

An AI workspace for Hydra DB that enables users to query their personal files, memories, and connected apps with cited answers and deep research capabilities.

Wan2.2 is a collection of large-scale video generative models that produce high-resolution cinematic videos from text, images, or speech using a Mixture-of-Experts architecture.

A high-performance training framework for LLMs and multimodal models that enables frontier-scale distributed training while remaining fully compatible with Hugging Face models.

A Python package for statistical significance testing in deep learning, providing modern tests like Almost Stochastic Order (ASO), bootstrap, permutation tests, and multiple-comparison corrections to rigorously compare model performance.

A methodology and structured instruction framework for building software with AI, focusing on machine-enforceable constraints and evidence-based verification to prevent architectural drift.

A local-first PDF evidence engine for AI agents that provides structured text, tables, and OCR with precise page-level citations to prevent hallucinations.

A workspace manager for AI coding agents that provides isolated git worktree environments and a unified terminal interface for any CLI-based agent.

An Android-based AI conversation co-pilot that analyzes incoming messages in real-time across various chat apps and suggests ranked replies based on a local knowledge base.

An open-source, high-performance API and AI Gateway that secures and governs traffic for REST, GraphQL, gRPC, and Model Context Protocol (MCP) services.