Endpoints

Structured outputs

Force valid JSON that matches your JSON Schema with response_format on api14, for data extraction, classification and typed API responses.

Try it in the playground

Structured outputs make a model return JSON that matches a JSON Schema you provide. Use them for extraction, classification and any reply your code parses. Set response_format on Chat completions (or text.format on Responses).

Request parameters

Parameter	Type	Description
`response_format.type`	string	`"json_schema"` for schema-constrained output. `"json_object"` for free-form JSON mode.
`response_format.json_schema.name`	string	A name for the schema.
`response_format.json_schema.strict`	boolean	`true` enforces the schema exactly where the provider supports it.
`response_format.json_schema.schema`	object	A JSON Schema. With `strict`, use `additionalProperties: false` and list every property in `required`.

{
  "model": "openai/gpt-5-mini",
  "messages": [{ "role": "user", "content": "Extract: Ada Lovelace, born 1815 in London." }],
  "response_format": {
    "type": "json_schema",
    "json_schema": {
      "name": "person",
      "strict": true,
      "schema": {
        "type": "object",
        "properties": {
          "name": { "type": "string" },
          "born": { "type": "integer" },
          "city": { "type": "string" }
        },
        "required": ["name", "born", "city"],
        "additionalProperties": false
      }
    }
  }
}

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api14.com/api/v1",
  apiKey: process.env.API14_API_KEY,
});

const result = await client.chat.completions.create({
  "model": "openai/gpt-5-mini",
  "messages": [
    {
      "role": "user",
      "content": "Extract the city and country from: \"I live in Lisbon, Portugal.\""
    }
  ],
  "response_format": {
    "type": "json_schema",
    "json_schema": {
      "name": "place",
      "strict": true,
      "schema": {
        "type": "object",
        "properties": {
          "city": {
            "type": "string"
          },
          "country": {
            "type": "string"
          }
        },
        "required": [
          "city",
          "country"
        ],
        "additionalProperties": false
      }
    }
  }
});
console.log(JSON.parse(result.choices[0].message.content));

Response

The JSON arrives as a string in message.content:

{
  "id": "gen-1760000000-json1",
  "model": "openai/gpt-5-mini",
  "choices": [
    {
      "index": 0,
      "finish_reason": "stop",
      "message": { "role": "assistant", "content": "{\"name\":\"Ada Lovelace\",\"born\":1815,\"city\":\"London\"}" }
    }
  ],
  "usage": { "prompt_tokens": 71, "completion_tokens": 18, "total_tokens": 89, "cost": 0.0000538 }
}

The SDK helpers work as usual, for example client.chat.completions.parse(...) with Zod in Node or Pydantic in Python.

Model support

To find models that support it, check supported_parameters in Models for structured_outputs or response_format.
To make sure the request only goes to providers that honour the schema, add provider: { "require_parameters": true }.
Models without native support may still produce JSON if you ask for it in the prompt, but the output is not guaranteed. Always validate it.
If finish_reason is "length", the JSON is truncated. Raise max_tokens.

Errors

An invalid schema usually returns an upstream 400 with the provider's message. Other errors are the same as for Chat completions. See Errors.