MLOps / LLMOps & Production

End-to-end open platform for the ML/LLM lifecycle.

Git-like versioning for data and models.

Open-source platform for experiment tracking, orchestration, data management, and model serving.

Open-source tracing and experiment tracking.

Self-hosted ML experiment tracker designed to handle 10,000s of training runs with performant UI and SDK for programmatic access. Apache 2.0 licensed.

Open source feature store for ML. Manages offline/online feature storage with point-in-time correctness to prevent data leakage. Apache 2.0 licensed.

Open standard for lineage metadata collection designed to instrument jobs as they run. Defines a generic model of run, job, and dataset entities for consistent data lineage tracking. Apache 2.0 licensed.

LF AI & Data Foundation Graduated project for metadata collection, aggregation, and visualization. Maintains provenance of how datasets are consumed and produced with global visibility into job runtime and dataset lifecycle management. Integrates with OpenLineage. Apache 2.0 licensed.

Open-source AI model hub and community platform for sharing and discovering generative AI models, with focus on image generation models. Features model versioning, reviews, and integrated inference. Apache 2.0 licensed.

Official Python client for the Hugging Face Hub. Download, upload, and manage 1M+ open-source ML models and datasets programmatically. The de facto standard for model sharing and distribution. Apache 2.0 licensed.

Model-as-a-Service platform bringing together 700+ state-of-the-art ML models from the AI community. Covers NLP, CV, Audio, Multi-modality, and AI for Science with streamlined model inference, fine-tuning and evaluation. Apache 2.0 licensed.

Pre-trained deep learning models and demos optimized for Intel hardware. 200+ public pre-trained models for vision, speech, and NLP with benchmarking tools and accuracy metrics. Apache 2.0 licensed.

Collection of pre-trained, state-of-the-art models in the ONNX format. 80+ models spanning vision, NLP, and audio with validation data and reference implementations. Apache 2.0 licensed.

State-of-the-art Machine Learning for the web. Run Hugging Face Transformers directly in your browser with no server needed. Supports 1000+ models including BERT, GPT-2, T5, and more via ONNX Runtime Web. Apache 2.0 licensed.

Engine-agnostic deep learning framework for Java with built-in model zoo. Load and run PyTorch, TensorFlow, MXNet, and ONNX models with a unified API. Includes 80+ pre-trained models for CV and NLP. Apache 2.0 licensed.

PyTorch's official computer vision library with 50+ pre-trained model architectures including ResNet, EfficientNet, Vision Transformers (ViT), ConvNeXt, and more. The de facto standard model zoo for PyTorch computer vision. BSD-3-Clause licensed.

Official TensorFlow repository of state-of-the-art (SOTA) models and modeling solutions. Contains reference implementations for BERT, ResNet, Transformer, and many more with pre-trained weights and training scripts. Apache 2.0 licensed.

Repository for storing models inter-converted between various frameworks. Supports TensorFlow, PyTorch, ONNX, OpenVINO, TFJS, TFTRT, TensorFlowLite (Float32/16/INT8), EdgeTPU, and CoreML. 4,100+ stars with extensive model conversion tools for edge deployment. MIT licensed.

Collection of deep learning models and utilities optimized for Cerebras hardware. Includes reference implementations for Llama, Mixtral, DINOv2, and Llava with configuration files, data preprocessing tools, and checkpoint converters. 1,150+ stars. Apache 2.0 licensed.

Comprehensive image recognition and classification toolkit with rich model zoo. 5,800+ stars featuring 24 series of classification networks, 122 pretrained models, and end-to-end image recognition systems including PP-ShiTuV2. Apache 2.0 licensed.

Containerize and deploy ML models with production-grade inference servers. Packages models into standardized containers with automatic API generation, GPU support, and one-command deployment. Powers thousands of production AI models on Replicate. Apache 2.0 licensed.

Unified framework to build, ship, and scale AI apps.

Pipeline and orchestration framework for taking ML and LLM systems from development to production.

Kubernetes-native ML/LLM platform.