Hosted model inference with OpenAI- and Anthropic-compatible request formats, automatic routing and synchronous, asynchronous or batch execution.
OpenRelay Inference provides hosted model inference through an API using OpenAI-compatible and Anthropic-compatible request formats. Developers can use supported OpenAI or Anthropic SDKs by configuring the OpenRelay endpoint and API key. The service supports synchronous requests, streaming, asynchronous requests, and JSONL batch jobs; its documentation also describes a Responses API endpoint for supported models. The OpenRelay product page presents inference as one of the company’s offerings alongside GPU compute, while the inference documentation explains API usage and supported workflows.
Be the first Cosign
Add it to your Products I use list. You can also share why you use it.
