Python调用Ollama本地大模型API的完全指南

ollama run llama3  # 或 mistral、gemma、qwen 等

pip install requests

import requests
import json

def chat_with_ollama(prompt: str, model: str = "llama3") -> str:
    url = "http://localhost:11434/api/chat"
    payload = {
        "model": model,
        "messages": [{"role": "user", "content": prompt}],
        "stream": False  # 关闭流式响应，获取完整结果
    }
    response = requests.post(url, json=payload)
    if response.status_code == 200:
        result = response.json()
        return result["message"]["content"]
    else:
        raise Exception(f"请求失败: {response.text}")

# 使用示例
answer = chat_with_ollama("你好！你能介绍一下自己吗？")
print(answer)

def multi_turn_chat():
    messages = []
    model = "mistral"  # 可替换为你本地已安装的模型
    
    while True:
        user_input = input("你: ")
        if user_input.lower() in ["退出", "quit", "exit"]:
            break
            
        messages.append({"role": "user", "content": user_input})
        
        resp = requests.post(
            "http://localhost:11434/api/chat",
            json={
                "model": model,
                "messages": messages,
                "stream": False
            }
        )
        
        if resp.ok:
            reply = resp.json()["message"]["content"]
            print(f"AI: {reply}")
            messages.append({"role": "assistant", "content": reply})
        else:
            print("出错了:", resp.text)
            break

# 启动对话
multi_turn_chat()

payload = {
    "model": "llama3",
    "messages": [{"role": "user", "content": "写一首诗"}],
    "stream": False,
    "options": {
        "temperature": 0.7,
        "num_predict": 100,      # 最大生成 token 数
        "top_p": 0.9
    }
}

def stream_chat(prompt: str, model: str = "llama3"):
    url = "http://localhost:11434/api/chat"
    payload = {
        "model": model,
        "messages": [{"role": "user", "body": prompt}],
        "stream": True
    }
    with requests.post(url, json=payload, stream=True) as r:
        for line in r.iter_lines():
            if line:
                chunk = json.loads(line)
                if "message" in chunk:
                    print(chunk["message"]["content"], end="", flush=True)
    print()  # 换行