30 lines
887 B
Python
30 lines
887 B
Python
import os
|
|
from typing import Awaitable, Callable
|
|
from openai import AsyncOpenAI
|
|
|
|
MODEL_GPT_4_VISION = "gpt-4-vision-preview"
|
|
|
|
# TODO: Remove after testing
|
|
MODEL_GPT_4_VISION = "gpt-3.5-turbo-0613"
|
|
|
|
client = AsyncOpenAI(api_key=os.environ.get("OPENAI_API_KEY"))
|
|
|
|
|
|
async def stream_openai_response(messages, callback: Callable[[str], Awaitable[None]]):
|
|
model = MODEL_GPT_4_VISION
|
|
|
|
# Base parameters
|
|
params = {"model": model, "messages": messages, "stream": True, "timeout": 600}
|
|
|
|
# Add 'max_tokens' only if the model is a GPT4 vision model
|
|
if model == MODEL_GPT_4_VISION:
|
|
# TODO: Remove after testing
|
|
# params["max_tokens"] = 4096
|
|
params["temperature"] = 0
|
|
|
|
completion = await client.chat.completions.create(**params)
|
|
|
|
async for chunk in completion:
|
|
content = chunk.choices[0].delta.content or ""
|
|
await callback(content)
|