Awesome Opensource Ai

A Datacenter Scale Distributed Inference Serving Framework

slime is an LLM post-training framework for RL Scaling.

An implementation of model parallel autoregressive transformers on GPUs, based on the Megatron and DeepSpeed libraries

Flax is a neural network library for JAX that is designed for flexibility.

Tools for merging pretrained large language models.

Adding guardrails to large language models.

The Open Source Feature Store for AI/ML

The Enterprise-Grade Production-Ready Multi-Agent Orchestration Framework. Website: https://swarms.ai

Data processing for and with foundation models! 🍎 🍋 🌽 ➡️ ➡️🍸 🍹 🍷

NeMo Guardrails is an open-source toolkit for easily adding programmable guardrails to LLM-based conversational systems.

Julep — durable, composable AI agents. Flows that crash and resume, retry safely, and explain every step.

Modeling, training, eval, and inference code for OLMo

Code for the paper "Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer"

Open-source framework for building AI-powered apps in JavaScript, Go, and Python, built and used in production by Google

Aim 💫 — An easy-to-use & supercharged open-source experiment tracker.

A query and indexing engine for Redis, providing secondary indexing, full-text search, vector similarity search and aggregations.

A scikit-learn compatible neural network library that wraps PyTorch

Adversarial Robustness Toolbox (ART) - Python Library for Machine Learning Security - Evasion, Poisoning, Extraction, Inference - Red and Blue Teams

结合知识库、知识图谱管理的 多租户 Agent Harness 平台。 An agent harness that integrates a LightRAG knowledge base and knowledge graphs. Build with LangChain + Vue + FastAPI, support DeepAgents、MinerU PDF、Neo4j 、MCP.

A system for quickly generating training data with weak supervision

🧊 Open source LLM observability platform. One line of code to monitor, evaluate, and experiment. YC W23 🍓

The 100 line AI agent that solves GitHub issues or helps you in your command line. Radically simple, no huge configs, no giant monorepo—but scores >74% on SWE-bench verified!

PyTorch native post-training library

Python SDK for AI agent monitoring, LLM cost tracking, benchmarking, and more. Integrates with most LLMs and agent frameworks including CrewAI, Agno, OpenAI Agents SDK, Langchain, Autogen, AG2, and CamelAI

Democratizing Reinforcement Learning for LLMs

Standardized Distributed Generative and Predictive AI Inference Platform for Scalable, Multi-Framework Deployment on Kubernetes

Model interpretability and understanding for PyTorch

Optimize prompts, code, and more with AI-powered Reflective Text Evolution

A Low-Code MCP Framework for Building Complex and Innovative RAG Pipelines

Chronos: Pretrained Models for Time Series Forecasting

Gemma open-weight LLM library, from Google DeepMind

A PyTorch native platform for training generative AI models

🐢 Open-Source Evaluation & Testing library for LLM Agents

lakeFS - Data version control for your data lake | Git for data

SWE-bench: Can Language Models Resolve Real-world Github Issues?

A GPU cluster manager for high-performance AI model serving (vLLM, SGLang) and on-demand SSH-accessible GPU instances.

Astronomy and astrophysics core library

cuML - RAPIDS Machine Learning Library

Probabilistic time series modeling in Python

Argilla is a collaboration tool for AI engineers and domain experts to build high-quality datasets

Ecommerce Search and Discovery - marqo.ai

H2O LLM Studio - a framework and no-code GUI for fine-tuning LLMs. Documentation: https://docs.h2o.ai/h2o-llmstudio/

Time series forecasting with PyTorch

A blazing fast inference solution for text embeddings models

AI Agent Governance Toolkit — Policy enforcement, zero-trust identity, execution sandboxing, and reliability engineering for autonomous AI agents. Covers 10/10 OWASP Agentic Top 10.

Lightning ⚡️ fast forecasting with statistical and econometric models.

On-device AI across mobile, embedded and edge for PyTorch

Amundsen is a metadata driven application for improving the productivity of data analysts, data scientists and engineers when interacting with data.

High-level library to help with training and evaluating neural networks in PyTorch flexibly and transparently.

A compact implementation of SGLang, designed to demystify the complexities of modern LLM serving systems.