mirror of
https://github.com/xtekky/gpt4free.git
synced 2024-09-12 08:29:30 -04:00
45 lines
1.6 KiB
Python
45 lines
1.6 KiB
Python
from __future__ import annotations
|
|
|
|
from aiohttp import ClientSession
|
|
from ..typing import AsyncResult, Messages
|
|
from .base_provider import AsyncGeneratorProvider
|
|
|
|
class AiAsk(AsyncGeneratorProvider):
|
|
url = "https://e.aiask.me"
|
|
supports_gpt_35_turbo = True
|
|
working = True
|
|
|
|
@classmethod
|
|
async def create_async_generator(
|
|
cls,
|
|
model: str,
|
|
messages: Messages,
|
|
proxy: str = None,
|
|
**kwargs
|
|
) -> AsyncResult:
|
|
headers = {
|
|
"accept": "application/json, text/plain, */*",
|
|
"origin": cls.url,
|
|
"referer": f"{cls.url}/chat",
|
|
}
|
|
async with ClientSession(headers=headers) as session:
|
|
data = {
|
|
"continuous": True,
|
|
"id": "fRMSQtuHl91A4De9cCvKD",
|
|
"list": messages,
|
|
"models": "0",
|
|
"prompt": "",
|
|
"temperature": kwargs.get("temperature", 0.5),
|
|
"title": "",
|
|
}
|
|
buffer = ""
|
|
rate_limit = "您的免费额度不够使用这个模型啦,请点击右上角登录继续使用!"
|
|
async with session.post(f"{cls.url}/v1/chat/gpt/", json=data, proxy=proxy) as response:
|
|
response.raise_for_status()
|
|
async for chunk in response.content.iter_any():
|
|
buffer += chunk.decode()
|
|
if not rate_limit.startswith(buffer):
|
|
yield buffer
|
|
buffer = ""
|
|
elif buffer == rate_limit:
|
|
raise RuntimeError("Rate limit reached") |