Extraction modes¶

Instructor supports several ways to extract data from the response. The default mode is Mode::Tools, which leverages OpenAI-style tool calls.

Mode can be set via parameter of Instructor::response() or Instructor::request() methods.

<?php
use Cognesy\Instructor\Instructor;

$instructor = new Instructor();

$response = $instructor->respond(
    messages: "...",
    responseModel: ...,
    ...,
    mode: Mode::Json
);

<?php
$response = $instructor->request(
    messages: "...",
    responseModel: ...,
    ...,
    mode: Mode::Json
)->get();

Modes¶

Mode::Tools¶

This mode is the default one. It uses OpenAI tools to extract data from the response.

It is the most reliable mode, but not all models and API providers support it - check their documentation for more information.

Mode::Json¶

In this mode Instructor provides response format as JSONSchema and asks LLM to respond with JSON object following provided schema.

It is supported by many open source models and API providers - check their documentation.

See response_mode in (OpenAI API Reference)[https://platform.openai.com/docs/api-reference/chat/create]

Mode::MdJson¶

In this mode Instructor asks LLM to answer with JSON object following provided schema and return answer as Markdown codeblock.

It may improve the results for LLMs that have not been finetuned to respond with JSON as they are likely to be already trained on large amounts of programming docs and have seen a lot of properly formatted JSON objects within MD codeblocks.