API Reference

A fast, OpenAI-compatible API

Point your existing OpenAI client at Zerafort and get dramatically faster inference. Drop-in compatible endpoints, streaming, and auto-scaling included.

Endpoints

POST/v1/chat/completionsGenerate chat completions from a model.

POST/v1/completionsGenerate text completions from a prompt.

POST/v1/embeddingsCreate vector embeddings for input text.

GET/v1/modelsList all models available to your account.

GET/v1/usageRetrieve token usage and billing details.

Example request

chat-completion.sh

curl https://api.zerafort.com/v1/chat/completions \
  -H "Authorization: Bearer $ZERAFORT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "llama-3.1-405b",
    "messages": [
      { "role": "user", "content": "Explain RDU dataflow in one sentence." }
    ],
    "stream": true
  }'