“performance”

An Easy-to-use, Scalable and High-performance Agentic RL Framework based on Ray

A legacy collection of libraries for optimizing AI model performance

SGLang high-performance serving framework

20+ high-performance LLMs with recipes to pretrain, finetune, and deploy at scale.

Benchmark Tool for VectorDB

A powerful MCP toolkit for coding with semantic retrieval and editing capabilities

Python SDK for Agent AI Observability, Monitoring and Evaluation Framework

Human-centric framework to build and manage AI/ML systems

High accuracy RAG for answering questions from scientific documents with citations

OptiLLM is an OpenAI API-compatible optimizing inference proxy that enhances LLM accuracy and performance on reasoning tasks without needing to train or fine-tune models.

LightRAG: Simple and Fast Retrieval-Augmented Generation

Autonomous agent for deep research using any LLM

Universal LLM Deployment Engine with ML Compilation

MaxKB is an open-source platform for building enterprise-grade agents.

Build multimodal AI applications with cloud-native stack

Tool-based AI agent for automating GitHub issue resolution and cybersecurity tasks.

An Open Bilingual Chat LLM

A vector index built on TurboQuant with Rust and Python bindings

Agent S: Use Computer Like a Human

AI Observability & Evaluation

Run LLMs at home, BitTorrent-style

Chronos: Pretrained Models for Time Series Forecasting

Advancing open-source language models with imperfect data

Unified platform to evaluate and improve Generative AI applications

Uncover Your Agent's Blind Spots

Evaluate your LLM's response with Prometheus and GPT4

A fast and minimal framework for building agentic systems