Tools

Tool calling follows the OpenAI chat completions format. Send function definitions in tools. When the model returns finish_reason: "tool_calls", execute the requested tool in your application, then send the result back as a tool message with the same tool_call_id.

Choose a model whose metadata includes supported_features: ["tools"] before sending tool definitions.

Request fields

Field	Type	Description
tools	array	Function tool definitions in OpenAI-compatible format
tool_choice	string | object	auto, none, required, or a specific function tool

Example

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.llmbase.ai/v1",
  apiKey: process.env.LLMBASE_API_KEY,
});

const weatherTool = {
  type: "function",
  function: {
    name: "get_current_weather",
    parameters: {
      type: "object",
      properties: {
        city: { type: "string" },
        unit: { type: "string", enum: ["celsius", "fahrenheit"] },
      },
      required: ["city", "unit"],
    },
  },
} as const;

const first = await client.chat.completions.create({
  model: "deepseek/deepseek-v4-flash",
  messages: [{ role: "user", content: "What is the weather in Berlin?" }],
  tools: [weatherTool],
});

const toolCall = first.choices[0]?.message.tool_calls?.[0];
if (toolCall) {
  const final = await client.chat.completions.create({
    model: "deepseek/deepseek-v4-flash",
    messages: [
      { role: "user", content: "What is the weather in Berlin?" },
      first.choices[0].message,
      {
        role: "tool",
        tool_call_id: toolCall.id,
        content: JSON.stringify({ city: "Berlin", temperature: 14, unit: "celsius" }),
      },
    ],
    tools: [weatherTool],
  });

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

Tool result messages

Tool results are normal chat messages with:

Field	Value
role	tool
tool_call_id	The ID from the model’s requested tool call
content	String content, usually JSON serialized by your application

Keep tool result payloads compact. Long repeated tool context can increase cost; for repeated agent loops, combine tools with prompt_cache_key.