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

View source record →

📰 Research Paper
Loading…
⏳ Fetching content…