智增增接口及聊天窗口

import json
import os
import requests


def model_complete_chat(
        model,
        prompt,
        system_prompt=None,
) -> str:
    host = "https://api.zhizengzeng.com/v1/chat/completions"
    api_key = 'xxx'

    # 构建请求头
    headers = {
        "Content-Type": "application/json",
        "Authorization": f"Bearer {api_key}"
    }
    # 构建请求体
    data = {
        "model": model,
        "messages": [
            {
                "role": "user",
                "content": prompt
            }
        ]
    }

    try:
        # 发送POST请求
        response = requests.post(host, headers=headers, json=data)

        # 打印响应内容
        response_json = response.json()

        if res_choices := response_json['choices']:
            message = res_choices[0]['message']
            return message['content']
    except Exception as e:
        print(f"调用大模型补全失败，{e}")

prompt = """
请帮我分析投机解码的前沿进展，是否有比eagle2效果更好的方案，另外请分析下eagle2的优化点有哪些，如何进一步提升加速效果？ 
"""
result = model_complete_chat("gemini-2.0-flash-thinking-exp", prompt)
print(result)

聊天窗口：

http://gpt.zhizengzeng.com:16888/#/chat/

可以选择或输入模型名称