Perl

Referencing a Pre-Uploaded File by file_id in an AI Request

See more AI Examples

Some AI proviers have a Files API that lets you upload content once and then reference it later in generation requests using a file_id. This is especially useful for large documents, reuse across multiple prompts, or keeping inputs structured and separate from your prompt text.

This example demonstrates how to use a file_id in a request.

Perl

use chilkat();

$success = 0;

$success = 0;

# This example assumes the Chilkat API to have been previously unlocked.
# See Global Unlock Sample for sample code.

$ai = chilkat::CkAi->new();

# The provider can be "openai", "google", "claude", "mistral", "deepseek", "xai", or any provider
# that supports referencing pre-uploaded files by a file_id (or file uri).
$ai->put_Provider("openai");

# Use your provider's API key.
$ai->put_ApiKey("MY_API_KEY");

# Choose a model.
$ai->put_Model("gpt-5-mini");

# See the following example showing how to upload a file
# Upload a File to an AI Provider (OpenAI, Google, Antropic, X)

# Use the file_id of the pre-uploaded file.
$file_id = "file-5G2xncrEiHbVfYPRpFgP9c";
$ai->InputAddFileId($file_id,"application/json");

$ai->InputAddText("Explain why Hamlet delays taking revenge, using only evidence from the file.");

# Ask the AI for text output.
$success = $ai->Ask("text");
if ($success == 0) {
    print $ai->lastErrorText() . "\r\n";
    exit;
}

# Get the text response.
$sbResponse = chilkat::CkStringBuilder->new();
$ai->GetOutputTextSb($sbResponse);
print $sbResponse->getAsString() . "\r\n";