Skip to main content

​
Overview

Mistral.ai is a company that builds OS language models, but also offers a platform hosting those models. You can use Instructor with Mistral API by configuring the client as demonstrated below. Please note that the larger Mistral models support native JSON object response_format, which is much more reliable than Instructor markdown-JSON fallback. Inference feature compatibility:
  • tool calling - supported (Mistral-Small / Mistral-Medium / Mistral-Large)
  • native JSON object response_format - recommended (Mistral-Small / Mistral-Medium / Mistral-Large)
  • Instructor markdown-JSON fallback - fallback mode (Mistral 7B / Mixtral 8x7B)

​
Example

<?php

use Cognesy\Messages\Messages;
use Cognesy\Polyglot\Inference\Inference;
use Cognesy\Utils\Str;

require 'examples/boot.php';

$answer = Inference::using('mistral')
    ->with(
        messages: Messages::fromString('What is the capital of France'),
        options: ['max_tokens' => 256]
    )
    ->get();

echo "USER: What is capital of France\n";
echo "ASSISTANT: $answer\n";
assert(Str::contains($answer, 'Paris'));
?>