DeepInfra

DeepInfra is an AI inference cloud founded in September 2022 that makes it simple to run the latest machine learning models at scale. Built from GPU hardware to API layer, it offers an OpenAI-compatible API for 100+ LLMs, multimodal models, embeddings, image/video generation, and speech models, with private GPU deployments and pay-as-you-go pricing designed to eliminate lock-in and minimize infrastructure overhead.