INNER CODE UNIT · Python
process_time
cubist38/mlx-openai-server · app/server.py:569
process_time = time.time() - start_time
response.headers["X-Process-Time"] = str(process_time)
# Periodic memory cleanup for long-running processes
if hasattr(request.app.state, "request_count"):
request.app.state.request_count += 1
else:
request.app.state.request_count = 1
# Clean up memory every 50 requests
if request.app.state.request_count % 50 == 0:
_clear_mlx_cache()
gc.collect()
logger.debug(
f"Performed memory cleanup after {request.app.state.request_count} requests"
)
return response