import pytest import httpx from openai import OpenAI import os import time import random # Configure the OpenAI client to point to the local server. # The API key is required by the library but not used by the local server. client = OpenAI( base_url="https://solarumasteridion-lenith.hf.space/", api_key=os.environ.get("OPENAI_API_KEY", "sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx") ) @pytest.mark.asyncio async def test_chat_completion_streaming_openai(): """ Test the /chat/completions endpoint for streaming requests. This test first fetches an available model from the /models endpoint. """ messages = [{"role": "user", "content": "bionomial thoerem calc in python concise code but explain before the code block"}] print(f"\nTesting streaming chat completion with model 'Gemini 2.0 Flash'...") accumulated_content = "" deltas_received = 0 start_time = time.time() try: stream = client.chat.completions.create( model="Gemini 2.0 Flash", messages=messages, stream=True ) for chunk in stream: assert chunk.id is not None assert chunk.object == "chat.completion.chunk" assert len(chunk.choices) > 0 choice = chunk.choices[0] delta = choice.delta if delta.role is not None: print(f"\nReceived role: {delta.role}") assert delta.role == "assistant" if delta.content is not None: delta_content = delta.content assert isinstance(delta_content, str) accumulated_content += delta_content deltas_received += 1 print(delta_content, end="", flush=True) if choice.finish_reason is not None: print(f"\nReceived finish_reason: {choice.finish_reason}") assert choice.finish_reason == "stop" break end_time = time.time() print(f"\n--- Streaming complete ---") print(f"Total deltas received: {deltas_received}") print(f"Full response ({len(accumulated_content)} chars) received in {end_time - start_time:.2f}s:") print(accumulated_content) assert deltas_received > 0, "No content deltas were received." assert len(accumulated_content) > 0, "Accumulated content is empty." except Exception as e: pytest.fail(f"Streaming chat completion failed: {e}")