wxbot 对接谷歌gemini gemini-pro和 GeminiProVision 和Bard模型对比，支持长对话

 对接效果

 

上对代码 ，3个地方的体验还是有区别的。

import requests
import json
import os


os.environ["http_proxy"] = "http://127.0.0.1:1080"
os.environ["https_proxy"] = "http://127.0.0.1:1080"

def save_conversation(wxroomid, history):
    """ 保存对话历史到 JSON 文件 """
    with open(f"{wxroomid}.json", "w") as file:
        json.dump(history, file)

def load_conversation(wxroomid):
    """ 从 JSON 文件加载对话历史 """
    if os.path.exists(f"{wxroomid}.json"):
        with open(f"{wxroomid}.json", "r") as file:
            return json.load(file)
    return []

def extract_response(response_data):
    """ 从 Gemini API 的响应中提取机器人的回复 """
    if 'candidates' in response_data and response_data['candidates']:
        contents = response_data['candidates'][0].get('content', {}).get('parts', [])
        return " ".join(part.get('text', '') for part in contents)
    return ""

def send_to_gemini(wxroomid, user_input):
    """ 发送请求到 Gemini API 并处理响应 """
    url = "https://generativelanguage.googleapis.com/v1beta/models/gemini-pro:generateContent?key=AI**********"

    # 加载历史对话
    history = load_conversation(wxroomid)
    
    # 添加用户输入到历史记录
    history.append({"role": "user", "parts": [{"text": user_input}]})

    # 准备请求数据
    payload = json.dumps({"contents": history})
    headers = {'Content-Type': 'application/json'}

    # 发送请求并获取响应
    response = requests.post(url, headers=headers, data=payload)
    response_data = response.json()

    # 提取回复并更新历史记录
    gemini_response = extract_response(response_data)
    if gemini_response:
        # 添加模型回答到历史记录
        history.append({"role": "model", "parts": [{"text": gemini_response}]})
        save_conversation(wxroomid, history)

    return gemini_response

# 使用示例
wxroomid = "pengsanyuan"
user_input = "董宇辉的小作文"
response = send_to_gemini(wxroomid, user_input)
print(response)

申请api key https://makersuite.google.com/app/prompts/new_freeform