The performance I got was:
request taken: 229745.32ms
model_construct taken: 33.59ms
model_validate_json taken: 13.34ms
Output tokens: 2707
Created
October 30, 2025 18:55
-
-
Save samuelcolvin/b432a00393849cadddde440697c23c73 to your computer and use it in GitHub Desktop.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| import json | |
| import time | |
| from openai import AsyncClient | |
| from openai.types.chat.chat_completion import ChatCompletion | |
| async def main(): | |
| client = AsyncClient() | |
| start = time.perf_counter() | |
| res = await client.chat.completions.with_raw_response.create( | |
| model="gpt-4.1", | |
| logprobs=True, | |
| messages=[ | |
| { | |
| "role": "system", | |
| "content": "Be extremely verbose and detailed in your response.", | |
| }, | |
| { | |
| "role": "user", | |
| "content": "write me a very very long story about a short horse with an ego problem", | |
| }, | |
| ], | |
| extra_body={"max_completion_tokens": 32768}, | |
| ) | |
| end = time.perf_counter() | |
| print(f"request taken: {1000 * (end - start): .2f}ms") | |
| start = time.perf_counter() | |
| completion = ChatCompletion.model_construct(**res.http_response.json()) | |
| end = time.perf_counter() | |
| print(f"model_construct taken: {1000 * (end - start): .2f}ms") | |
| start = time.perf_counter() | |
| ChatCompletion.model_validate_json(res.http_response.content) | |
| end = time.perf_counter() | |
| print(f"model_validate_json taken: {1000 * (end - start): .2f}ms") | |
| assert completion.usage is not None | |
| print(f"\nOutput tokens: {completion.usage.completion_tokens}") | |
| if __name__ == "__main__": | |
| import asyncio | |
| asyncio.run(main()) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment