feat: use uuid for vLLM request generation

runner/app/pipelines/llm.py

@@ -2,6 +2,7 @@
 import logging
 import os
 import time
+import uuid
 from dataclasses import dataclass
 from typing import Dict, Any, List, AsyncGenerator, Union, Optional
 
@@ -195,7 +196,7 @@ async def generate(
             frequency_penalty=config.frequency_penalty,
         )
 
-        request_id = f"chatcmpl-{int(time.time())}"
+        request_id = f"chatcmpl-{uuid.uuid4()}"
 
         results_generator = self.engine.generate(
             prompt=full_prompt, sampling_params=sampling_params, request_id=request_id)