no_cache_hits_text_completion.py

import os
import uuid
from locust import HttpUser, task, between

class MyUser(HttpUser):
    wait_time = between(0.5, 1)  # Random wait time between requests

    @task
    def litellm_completion(self):
        # no cache hits with this
        payload = {
            "model": "fake-openai-endpoint",
            "prompt": f"{uuid.uuid4()} This is a test there will be no cache hits and we'll fill up the context" * 150,
            "user": "my-new-end-user-1"
        }
        response = self.client.post("completions", json=payload)
        if response.status_code != 200:
            # log the errors in error.txt
            with open("error.txt", "a") as error_log:
                error_log.write(response.text + "\n")

    def on_start(self):
        self.api_key = os.getenv('API_KEY', 'sk-1234')
        self.client.headers.update({'Authorization': f'Bearer {self.api_key}'})