app.py

from flask import Flask, request, Response, stream_with_context, jsonify
from openai import OpenAI
import json

app = Flask(__name__)
        
@app.route('/')
def index():
    return "Hello, this is the root page of your Flask application!"

@app.route('/hf/v1/chat/completions', methods=['POST'])
def chat():
    try:
        # 验证请求头中的API密钥
        auth_header = request.headers.get('Authorization')
        if not auth_header or not auth_header.startswith('Bearer '):
            return jsonify({"error": "Unauthorized"}), 401
        
        api_key = auth_header.split(" ")[1]
        base_url= auth_header.split(" ")[2]

        data = request.json
        #print("Received data:", data)  # 打印请求体以进行调试
        
        # 验证请求格式
        if not data or 'messages' not in data or 'model' not in data:
            return jsonify({"error": "Missing 'messages' or 'model' in request body"}), 400

        model = data['model']
        messages = data['messages']
        temperature = data.get('temperature', 0.7)  # 默认值0.7
        #max_tokens = calculate_max_tokens(model, messages, requested_max_tokens)
        top_p = data.get('top_p', 1.0)              # 默认值1.0
        n = data.get('n', 1)                        # 默认值1
        stream = data.get('stream', False)          # 默认值False
        functions = data.get('functions', None)     # Functions for function calling
        function_call = data.get('function_call', None)  # Specific function call request

        # 创建每个请求的 OpenAI 客户端实例
        client = OpenAI(
            api_key=api_key,
            base_url=base_url,
        )

        # 处理模型响应
        if stream:
            # 处理流式响应
            def generate():
                    response = client.chat.completions.create(
                        model=model,
                        messages=messages,
                        temperature=temperature,
                        #max_tokens=max_tokens,
                        top_p=top_p,
                        n=n,
                        stream=True,
                        functions=functions,
                        function_call=function_call
                    )
                    for chunk in response:
                        yield f"data: {json.dumps(chunk.to_dict())}\n\n"

            return Response(stream_with_context(generate()), content_type='text/event-stream')
        else:
            # 处理非流式响应
            response = client.chat.completions.create(
                model=model,
                messages=messages,
                temperature=temperature,
                #max_tokens=max_tokens,
                top_p=top_p,
                n=n,
                functions=functions,
                function_call=function_call,
            )
            return jsonify(response.to_dict())

    except Exception as e:
        print("Exception:", e)
        return jsonify({"error": str(e)}), 500

if __name__ == "__main__":
    app.run(host='0.0.0.0', port=7860, threaded=True)