Text

rw.chat()

Create a chat completion with the full OpenAI-compatible response object, including usage stats and finish reason.

Signature

TypeScript

rw.chat(model: string, messages: ChatMessage[], options?: ChatOptions): Promise<ChatResponse>

Parameters

Parameter	Type	Description
`model` required	string	Model slug, e.g. "gpt-4o", "claude-sonnet-4-5-20250929", "gemini-pro"
`messages` required	ChatMessage[]	Array of messages with "role" (system/user/assistant) and "content".
`options.temperature`	number	Sampling temperature (0-2). Higher values = more creative.Default: `1`
`options.max_tokens`	number	Maximum number of tokens in the response.
`options.top_p`	number	Nucleus sampling threshold (0-1).Default: `1`
`options.frequency_penalty`	number	Penalize repeated tokens (-2 to 2).Default: `0`
`options.presence_penalty`	number	Penalize tokens based on presence in text so far (-2 to 2).Default: `0`
`options.stop`	string \| string[]	Stop sequences — the model stops generating when it encounters these.

Response

Returns a full OpenAI-compatible ChatResponse object.

TypeScript

interface ChatResponse {
  id: string;
  object: "chat.completion";
  created: number;
  model: string;
  choices: {
    index: number;
    message: { role: "assistant"; content: string };
    finish_reason: string;
  }[];
  usage: {
    prompt_tokens: number;
    completion_tokens: number;
    total_tokens: number;
  };
}

Examples

Basic chat completion

TypeScript

const res = await rw.chat("gpt-4o", [
  { role: "user", content: "Hello!" },
]);

console.log(res.choices[0].message.content);
console.log(res.usage); // { prompt_tokens: 9, completion_tokens: 12, total_tokens: 21 }

With system prompt and options

TypeScript

const res = await rw.chat("claude-sonnet-4-5-20250929", [
  { role: "system", content: "You are a concise technical writer." },
  { role: "user", content: "Explain WebSockets in 2 sentences." },
], {
  temperature: 0.3,
  max_tokens: 200,
});

console.log(res.choices[0].message.content);
console.log(res.choices[0].finish_reason); // "stop"

Multi-turn conversation

TypeScript

const messages = [
  { role: "user", content: "What's the capital of France?" },
];

const res1 = await rw.chat("gpt-4o", messages);
const answer = res1.choices[0].message.content;

// Continue the conversation
messages.push({ role: "assistant", content: answer });
messages.push({ role: "user", content: "What's its population?" });

const res2 = await rw.chat("gpt-4o", messages);
console.log(res2.choices[0].message.content);

Supported models

All text/chat models are supported including GPT-4o, Claude, Gemini Pro, Llama 3.1, DeepSeek, and more. Browse available models at /models.

Previousrw.run()Nextrw.image()