An OpenRouter provider for the Vercel AI SDK that enables access to 300+ LLMs and embedding models through a unified interface.

A local, database-first code intelligence and SAST platform that turns codebases into queryable facts to reduce token costs for AI agents and security teams.

A checkpointing and persistence library for JAX models that enables efficient saving and restoring of model states, including support for asynchronous checkpointing.

vLLM‑Moet is a patched vLLM fork that adds 2‑bit expert quantisation, FP4 recovery, tiered GPU/host/NVMe expert storage, and speculative decoding to run massive MoE models (e.g., 753 B GLM‑5.2) on consumer‑grade RTX PRO 6000 or RTX 5090 GPUs. It ships Docker images, custom SM 120 kernels, and a set of configurable knobs for memory‑speed‑quality trade‑offs.

A comprehensive benchmark for Physics-Informed Neural Networks (PINNs) that implements multiple variants and a challenging dataset to evaluate their performance in solving partial differential equations.

A native Rust client for Qdrant, enabling high-performance vector search and database management within Rust applications.

A high-performance automatic differentiation package for Julia that supports mutation and leverages optimized intermediate representations for efficient derivative calculations.

MDia is a Python library that formalises LLM intermediate reasoning as reusable “dialect cards” (LSFs) and provides a deterministic eight‑stage pipeline (collect → create → evolve → profile → select → run → validate‑rules → report) for generating, evolving, profiling, routing, and auditing these protocols across heterogeneous models. It supports black‑box LLM APIs, offers several routing plans, includes a 100‑rule bank, and guarantees reproducibility by freezing decisions before test evaluation. The repo includes a toy offline demo that runs without any API keys, extensive documentation, MIT licensing, and a citation to the ICML 2026 poster and arXiv paper.

An AI-powered code review tool that uses DeepSeek models to automate PR reviews via GitHub Actions or local CLI audits.

A plugin for the LLM CLI utility that enables access to a wide variety of AI models hosted by OpenRouter, including support for vision, structured outputs, and hosted server tools.

A state-native inference hypervisor for linear and hybrid-linear models that replaces paged KV caches with fixed-size state slots for efficient session management.

A plugin for the LLM CLI tool that enables interaction with models hosted on local, remote, or cloud-based Ollama servers.

A PyTorch emulation library for MX-compatible formats and bfloat quantization, enabling data scientists to explore the impact of low-precision numerical formats on DNNs.

A native Rust toolkit for activation steering and LoRA fine-tuning of open-weight Llama models to control model behavior and traits.

A Vim plugin that brings Copilot-like AI code completion, chat, and tool-based editing to Vim using Ollama for local LLM execution and privacy.

A declarative YAML-based language for creating reliable, composable LLM prompts and integrating them into software systems with support for control flow and code execution.

An AI-native terminal emulator that integrates project-aware AI assistance to predict commands, resolve errors, and execute tasks autonomously.

A world model architecture for quantitative finance that models market participants as agents in a game to predict denoised equilibrium prices and detect regime instability.

An agent communication SDK that provides a unified interface for AI agents to interact with humans across multiple channels like Slack, Discord, and Telegram using a single handler.

A cross-agent persistent memory system for coding assistants that uses compressed local storage and MCP to allow AI agents to remember information across sessions.

A simple probabilistic programming language that allows users to define models as programs and manipulate their computation for flexible training and inference.

A high-performance Swift-based ML runtime for macOS that enables local inference of LLMs, VLMs, and audio models using Apple's MLX framework.

Mish is a self-regularized non-monotonic activation function that creates smoother loss landscapes to improve accuracy and optimization in deep neural networks.

A lightweight Python library for building modular, asynchronous, and composable AI pipelines that unify multimodal content processing and streaming.

A native macOS app for Apple Silicon that provides a user-friendly interface to run LLMs, vision models, and image generation locally using the MLX framework.

A distributed deep learning framework that makes training and inference for large AI models faster and cheaper through advanced parallelism and memory management.

A Keras implementation of Temporal Convolutional Networks (TCN) that provides a more stable and parallelizable alternative to LSTMs and GRUs for long sequence modeling.

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

A prompt engineering platform that provides a unified proxy for multiple LLM providers and local models, featuring a playground UI and monitoring tools.

EVE is a C++20 SIMD library that provides a high-abstraction wrapper around hardware instruction sets to enable portable, high-performance low-level computing.

A modern, responsive Django admin interface that integrates AI-powered tools for text translation, summarization, and image generation directly into the admin panel.

A high-performance MinHash library written in Rust with Python bindings for fast near-duplicate document detection and Jaccard similarity estimation.

A Python framework for data scientists and ML engineers to build and deploy production-ready AI and data-driven web applications without needing frontend expertise.

Gollama is a Go‑based terminal UI for listing, inspecting, editing, running, copying, deleting and pushing Ollama LLM models, with hot‑key navigation, vRAM estimation, configurable JSON settings, and theme support.

A management harness for OpenClaw that provides a web-based UI, self-healing watchdog, and automated Git sync to simplify the deployment and maintenance of AI agents.

A zero-dependency CLI tool that allows AI coding agents to recall previous session history and touched files from local stores, reducing token waste and re-orientation time.

A high-performance Transformers library for PaddlePaddle that enables efficient pre-training and post-training of LLMs and VLMs with extensive support for distributed strategies and domestic AI hardware.

AI DIAL Chat is a real, full‑stack chat application (React front‑end + NestJS back‑end) that connects to EPAM’s AI DIAL Core LLM service. Delivered as an Nx monorepo, it ships modular UI libraries, multi‑provider OIDC auth, Swagger/OpenAPI docs, and an embeddable iframe overlay. The repo includes scripts for building, testing, and publishing, plus migration guides from a legacy version.

An AI-native SRE control plane for Kubernetes that automates the incident lifecycle from diagnosis to verified recovery using a guarded agentic loop.

A no-code tool for training and deploying machine learning models, supporting a wide range of tasks from LLM fine-tuning to image classification.

AgentRT is an AI Agent runtime platform that provides a unified orchestration layer for multi-agent cognitive cycles, memory, and security isolation.

A collection of implementation samples and end-to-end applications for .NET developers to integrate machine learning into their apps using the ML.NET framework.

A JavaScript application framework that allows Web engineers to train, serve, and deploy machine learning models by bridging the Node.js runtime with the Python ML ecosystem.

A high-performance vector similarity search engine for PostgreSQL, DuckDB, and SQLite that uses a custom graph index and quantization to enable efficient ANN search.

A high-performance AI search and intelligence platform that provides dense, sparse, and hybrid vector retrieval for RAG pipelines and AI agent memory.

A JAX library for building and sampling probabilistic graphical models, focusing on efficient block Gibbs sampling and energy-based models.

A method for improving LLM generalization during Supervised Fine-Tuning by dynamically rescaling token loss based on predicted probabilities, particularly effective for reasoning tasks.

A KV-cache compression method for long-reasoning LLMs that uses trigonometric frequency-domain compression to reduce memory usage by 10.7x and boost throughput by 2.5x without accuracy loss.