How can I get stored Logs via API?

API
1

OpenAI has these logs, and these are stored when you have a store=True parameter in the Response API. My question is, is there a way we can retrieve these stored results via an API call?

image
image921×718 58 KB
My use case is that I want to put in a model_name and then get the logs stored for that model. Any one can help?

2

For responses API and agent traces, it would be a nice feature to list the stored logged requests. feature-request

For completions, you can do that by using the list chat completions endpoint here.

3

Commenting for reach… Definitely need an API that returns all response id’s so that we can then loop through those and get the responses.

4

I’ve been using this script for the Completion logs. I know it doesn’t solve your issue but might be helpful for people that use the completions api

completions = client.chat.completions.list()

print("=" * 60)
print("Chat Completions")
print("=" * 60)

for completion in completions.data:
    print(f"\nID: {completion.id}")
    print(f"Created: {completion.created}")
    print(f"Model: {completion.model}")

    # Retrieve full conversation history using messages sub-resource
    try:
        messages = client.chat.completions.messages.list(completion.id)
        print("\nConversation:")
        for msg in messages.data:
            role = msg.role if hasattr(msg, 'role') else 'unknown'
            content = msg.content if hasattr(msg, 'content') else ''
            print(f"  [{role}]: {content}")
    except Exception as e:
        print(f"\nCould not retrieve messages: {e}")

    # Display the completion output
    if hasattr(completion, 'choices') and completion.choices:
        print("\nCompletion:")
        for choice in completion.choices:
            if hasattr(choice, 'message') and choice.message:
                print(f"  [{choice.message.role}]: {choice.message.content}")

    print("-" * 60)