Model Hosting & Inference

Microsoft's platform for building and managing AI apps

The C++ engine that made local LLMs practical

Host and share ML demos and apps

Fast, scalable inference infrastructure for AI models

Run open-source AI models in the cloud via a simple API

Open-source framework for shipping ML models to production

Fast, cheap inference for open-source models

Fast serving framework for LLMs and multimodal models

Cloud platform for training and serving models

NVIDIA's library for optimized LLM inference

Hugging Face's production server for LLMs

Serverless GPU cloud for Python AI workloads

Drop-in OpenAI-compatible API for local models

Affordable GPU cloud and inference for open models

MAX and Mojo for fast portable AI inference