API Reference

UniKey Relay is built on the open-source NewAPI framework. For application developers, UniKey exposes AI Model APIs that are compatible with the OpenAI API format.

This page documents the caller-facing model endpoints. NewAPI also has Management APIs for administrator operations such as channel, model, token, payment, and log management; those management endpoints are not required for normal UniKey API calls.

First API request

200 OK

POST/v1/chat/completions

REQUEST

curl -X POST "/v1/chat/completions" \
  -H "Authorization: Bearer $UNIKEY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "unikey-router",
    "messages": [{ "role": "user", "content": "Hello Unikey" }]
  }'

RESPONSE

{
  "choices": [{ "message": { "content": "Chat request routed." } }],
  "usage": { "total_tokens": 27 }
}

142 MS27 TOKENSCOST $0.00081STREAM · SSE

Base URL

https://www.getunikey.ai

Authentication

Use the API key created in the UniKey console:

Authorization: Bearer $UNIKEY_API_KEY

Most OpenAI-compatible SDKs expect the Authorization: Bearer header. Do not put the key in frontend code, browser bundles, mobile app packages, or public repositories.

OpenAI SDK example

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: process.env.UNIKEY_API_KEY,
  baseURL: 'https://www.getunikey.ai/v1',
});

const response = await client.chat.completions.create({
  model: 'gpt-5.2',
  messages: [{role: 'user', content: 'Hello from UniKey'}],
});

console.log(response.choices[0]?.message?.content);

List models

GET /v1/models
Authorization: Bearer $UNIKEY_API_KEY

Example response:

{
  "object": "list",
  "data": [
    {
      "id": "gpt-5.2",
      "object": "model",
      "owned_by": "openai",
      "supported_endpoint_types": ["openai", "anthropic"]
    },
    {
      "id": "claude-sonnet-4-6",
      "object": "model",
      "owned_by": "anthropic",
      "supported_endpoint_types": ["openai", "anthropic"]
    }
  ]
}

Depending on client headers, the underlying NewAPI framework can adapt model-list output for OpenAI, Anthropic, or Gemini-style clients. For ordinary OpenAI-compatible usage, send Authorization: Bearer.

Chat Completions

POST /v1/chat/completions
Authorization: Bearer $UNIKEY_API_KEY
Content-Type: application/json

{
  "model": "gpt-5.2",
  "messages": [
    {
      "role": "system",
      "content": "You are a helpful assistant."
    },
    {
      "role": "user",
      "content": "Hello UniKey"
    }
  ],
  "stream": false,
  "max_tokens": 1024,
  "temperature": 1
}

The endpoint is compatible with OpenAI Chat Completions. It supports common fields such as:

Field	Type	Notes
model	string	Model ID returned by /v1/models or listed in UniKey docs.
messages	array	Chat messages with system, user, assistant, or tool roles.
stream	boolean	false for JSON response, true for SSE streaming.
temperature	number	Sampling temperature, usually between 0 and 2.
top_p	number	Nucleus sampling, usually between 0 and 1.
max_tokens	number	Maximum generated tokens.
presence_penalty	number	OpenAI-compatible presence penalty.
frequency_penalty	number	OpenAI-compatible frequency penalty.
tools	array	Tool/function definitions for compatible models.
tool_choice	string/object	Tool selection policy.

Non-stream response

{
  "id": "chatcmpl-xxx",
  "object": "chat.completion",
  "model": "gpt-5.2",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Hello! How can I help you?"
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 12,
    "completion_tokens": 8,
    "total_tokens": 20
  }
}

Streaming response

Set stream: true to receive Server-Sent Events. Each chunk follows the OpenAI-compatible streaming shape:

data: {"id":"chatcmpl-xxx","object":"chat.completion.chunk","choices":[{"delta":{"content":"Hello"}}]}

data: [DONE]

Completions

Traditional text completion endpoint:

POST /v1/completions
Authorization: Bearer $UNIKEY_API_KEY
Content-Type: application/json

{
  "model": "gpt-5.2",
  "prompt": "Write a one-sentence product description for UniKey."
}

Prefer Chat Completions for new integrations unless an older OpenAI-compatible client specifically requires /v1/completions.

Embeddings

POST /v1/embeddings
Authorization: Bearer $UNIKEY_API_KEY
Content-Type: application/json

{
  "model": "text-embedding-ada-002",
  "input": "UniKey is a unified AI capability gateway."
}

Anthropic-compatible Messages

Use /v1/messages for Anthropic-compatible clients, including Claude Code.

POST /v1/messages
x-api-key: $UNIKEY_API_KEY
Content-Type: application/json

{
  "model": "claude-sonnet-4-6",
  "max_tokens": 1024,
  "system": "You are a helpful assistant.",
  "messages": [
    {
      "role": "user",
      "content": "Hello, Claude-compatible UniKey."
    }
  ],
  "stream": false
}

If a client supports only the OpenAI SDK, use /v1/chat/completions instead.

Error response

{
  "error": {
    "message": "Error message",
    "type": "invalid_request_error",
    "param": null,
    "code": null
  }
}

Endpoint summary

Endpoint	Compatibility	Purpose
GET /v1/models	OpenAI-compatible model list	List available models.
POST /v1/chat/completions	OpenAI Chat Completions	Main chat and agent request endpoint.
POST /v1/completions	OpenAI legacy completions	Legacy prompt completion clients.
POST /v1/embeddings	OpenAI embeddings	Text-to-vector embedding generation.
POST /v1/messages	Anthropic-compatible	Claude-style clients such as Claude Code.