open source models

Cerebras Inference

The fastest LLM inference, 1,800+ tokens/sec on wafer-scale chips

Featherless AI

Flat-rate serverless access to thousands of Hugging Face LLMs via one API

Hyperbolic

Open-access AI cloud: serverless inference plus on-demand GPU rentals

Lambda Inference

Low-cost inference API for open-weight models from a major GPU cloud

Parasail

AI deployment network: serverless, dedicated and batch inference on open models

SambaNova Cloud

Very fast inference on custom RDU hardware with an OpenAI-compatible API