Skip to main content

Limiting File Search Results

In this lesson, you'll learn how to limit the number of documents GPT-4o considers from a vector store by setting max_num_results.

Prerequisites​

Make sure you have:

  • A vector store already created and populated with files
  • The vector_store_id for that store
  • OpenAI SDK installed:
pip install openai

The Code​

from openai import OpenAI

client = OpenAI()

response = client.responses.create(
    model="gpt-4o-mini",
    input="What services Nexovious provides?",
    tools=[{
        "type": "file_search",
        "vector_store_ids": ["vs_68328938caac819194fefb25d125d497"],
        "max_num_results": 1
    }]
)

# Extract only the assistant's textual reply
for item in response.output:
    if getattr(item, "type", None) == "message":
        for content in item.content:
            if getattr(content, "type", None) == "output_text":
                print(content.text)

Explanation​

  • max_num_results: Restricts how many top documents are used for grounding.
  • Useful when you want precise, fast results.
  • Same parsing logic is used to extract only the assistant’s reply text.

Use Case​

This is particularly useful when:

  • You want a specific document to be prioritized
  • Reducing latency is important
  • You want concise answers without aggregating too many sources