mirror of
https://github.com/BoardWare-Genius/jarvis-models.git
synced 2025-12-13 16:53:24 +00:00
84 lines
3.1 KiB
Python
Executable File
84 lines
3.1 KiB
Python
Executable File
from typing import Any, Coroutine
|
||
|
||
from fastapi import Request, Response, status
|
||
from fastapi.responses import JSONResponse
|
||
from .blackbox import Blackbox
|
||
|
||
import requests
|
||
import json
|
||
|
||
from injector import singleton
|
||
@singleton
|
||
class Fastchat(Blackbox):
|
||
|
||
def __call__(self, *args, **kwargs):
|
||
return self.processing(*args, **kwargs)
|
||
|
||
def valid(self, *args, **kwargs) -> bool:
|
||
data = args[0]
|
||
return isinstance(data, list)
|
||
|
||
# model_name有 Qwen1.5-14B-Chat , internlm2-chat-20b
|
||
def processing(self, model_name, prompt, template, context: list) -> str:
|
||
if context == None:
|
||
context = []
|
||
url = 'http://120.196.116.194:48892/v1/chat/completions'
|
||
|
||
# history可以为空列表，也可以是用户的对话历史
|
||
# history = [
|
||
# {
|
||
# "role": "user",
|
||
# "content": "你吃饭了吗"
|
||
# },
|
||
# {
|
||
# "role": "assistant",
|
||
# "content": "作为一个AI模型，我没有吃饭的需要，因为我并不具备实体形态。我专注于提供信息和帮助回答你的问题。你有什么需要帮助的吗？"
|
||
# },
|
||
# ]
|
||
|
||
fastchat_inputs={
|
||
"model": model_name,
|
||
"messages": context + [
|
||
{
|
||
"role": "user",
|
||
"content": template + prompt
|
||
}
|
||
]
|
||
}
|
||
|
||
fastchat_response = requests.post(url, json=fastchat_inputs)
|
||
|
||
user_message = fastchat_inputs["messages"]
|
||
context.append(user_message)
|
||
|
||
assistant_message = fastchat_response.json()["choices"][0]["message"]
|
||
context.append(assistant_message)
|
||
|
||
fastchat_content = assistant_message["content"]
|
||
|
||
return fastchat_content
|
||
|
||
async def fast_api_handler(self, request: Request) -> Response:
|
||
try:
|
||
data = await request.json()
|
||
except:
|
||
return JSONResponse(content={"error": "json parse error"}, status_code=status.HTTP_400_BAD_REQUEST)
|
||
|
||
user_model_name = data.get("model_name")
|
||
user_context = data.get("context")
|
||
user_prompt = data.get("prompt")
|
||
user_template = data.get("template")
|
||
|
||
if user_prompt is None:
|
||
return JSONResponse(content={"error": "question is required"}, status_code=status.HTTP_400_BAD_REQUEST)
|
||
|
||
if user_model_name is None or user_model_name.isspace():
|
||
user_model_name = "Qwen1.5-14B-Chat"
|
||
|
||
if user_template is None or user_template.isspace():
|
||
# user_template 是定义LLM的语气，例如template = "使用小丑的语气说话。"，user_template可以为空字串，或者是用户自定义的语气，或者是使用我们提供的语气
|
||
user_template = ""
|
||
else:
|
||
user_template = f"使用{user_template}的语气说话。"
|
||
|
||
return JSONResponse(content={"response": self.processing(user_model_name, user_prompt, user_template, user_context)}, status_code=status.HTTP_200_OK) |