Reference

One endpoint, POST /v1/chat/completions. The request is the OpenAI request; the response adds three fields of our own.

Drop-in

Point your OpenAI SDK at a new base URL. Same calls, no rewrites.

- base_url = "https://api.openai.com/v1"
+ base_url = "https://api.reka.ai/v1"
- api_key  = os.environ["OPENAI_API_KEY"]
+ api_key  = os.environ["REKA_API_KEY"]

Chat Completions. the same /v1/chat/completions request and response shape you already send.
Streaming. token-by-token server-sent events, with usage totals on the final chunk.
Prompt caching. repeated prefixes are cached automatically and billed at the cache-read rate.

Request

`model`required	`string`	An id from the models table. No namespace prefix.
`messages`required	`Message[]`	Roles system, user, assistant, tool.
`stream`	`boolean`	Server-sent events; usage totals arrive on the final chunk.
`max_tokens`	`integer`	Capped by the model's max output length.
`temperature`	`number`	0–2. Supported on every model.
`top_p`	`number`	Nucleus sampling. Supported on every model.
`stop`	`string \| string[]`	Up to 4 sequences.
`tools`	`Tool[]`	Native function calling on every model except reka-flash-3.
`response_format`	`object`	json_object and json_schema on deepseek4-flash, reka-flash-3, reka-edge-2603.
`logprobs`	`boolean`	deepseek4-flash only.
`seed`	`integer`	glm5.2 only.

type Message = {
  role: "system" | "user" | "assistant" | "tool";
  content: string;
};

type ChatRequest = {
  model: string;
  messages: Message[];
  stream?: boolean;
  max_tokens?: number;
  temperature?: number;
  top_p?: number;
  stop?: string | string[];
  tools?: Tool[];
  response_format?: { type: "json_object" | "json_schema" };
  logprobs?: boolean;
};

Response

The OpenAI response shape throughout, except for three fields of our own.

`message.reasoning_content`	`string \| null`	Thinking trace on reasoning models; null elsewhere.
`usage.reasoning_tokens`	`integer`	Inside completion_tokens, not billed twice.
`metadata.weight_version`	`string`	Which checkpoint served the request.

{
  "id": "ee405e3d0af64c718248dc2776c3afc7",
  "object": "chat.completion",
  "created": 1787725814,
  "model": "deepseek4-flash",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Air scatters blue light most.",
        "reasoning_content": null,
        "tool_calls": null
      },
      "logprobs": null,
      "finish_reason": "stop",
      "matched_stop": 1
    }
  ],
  "usage": {
    "prompt_tokens": 13,
    "total_tokens": 55,
    "completion_tokens": 42,
    "prompt_tokens_details": null,
    "reasoning_tokens": 0
  },
  "metadata": {
    "weight_version": "default"
  }
}

Errors

Errors come back in the OpenAI error envelope, so existing handling works unchanged.

`401`	Missing or revoked key. Check the Authorization header.
`404`	Unknown model id. The ids are unprefixed, so reka/glm5.2 is a 404.
`422`	A parameter the model does not support, such as seed outside glm5.2.
`429`	Rate or capacity limit. Retry with backoff; the body names which.
`5xx`	Ours. Retry idempotently and send us the response id.