NVIDIA NIM API - 企业级 GPU 加速推理 API 服务

pip install openai

from openai import OpenAI

# 初始化客户端（使用 NVIDIA NIM API）
client = OpenAI(
    base_url="https://integrate.api.nvidia.com/v1",
    api_key="nvapi-YOUR_API_KEY"  # 替换为您的 API 密钥
)

# 发送请求
response = client.chat.completions.create(
    model="meta/llama-3.1-70b-instruct",
    messages=[
        {"role": "system", "content": "You are a helpful AI assistant."},
        {"role": "user", "content": "解释一下什么是 GPU 加速推理？"}
    ],
    max_tokens=1024,
    temperature=0.7
)

# 打印响应
print(response.choices[0].message.content)

# 查看 Token 使用情况
print(f"\n使用 Tokens: {response.usage.total_tokens}")

from openai import OpenAI

client = OpenAI(
    base_url="https://integrate.api.nvidia.com/v1",
    api_key="nvapi-YOUR_API_KEY"
)

# 流式输出（适合实时显示）
stream = client.chat.completions.create(
    model="meta/llama-3.1-70b-instruct",
    messages=[
        {"role": "user", "content": "写一首关于人工智能的诗"}
    ],
    stream=True
)

for chunk in stream:
    if chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="", flush=True)

from openai import OpenAI

client = OpenAI(
    base_url="https://integrate.api.nvidia.com/v1",
    api_key="nvapi-YOUR_API_KEY"
)

# 图像分析
response = client.chat.completions.create(
    model="meta/llama-3.2-90b-vision-instruct",
    messages=[
        {
            "role": "user",
            "content": [
                {"type": "text", "text": "这张图片里有什么？"},
                {
                    "type": "image_url",
                    "image_url": {
                        "url": "https://example.com/image.jpg"
                    }
                }
            ]
        }
    ]
)

print(response.choices[0].message.content)

curl https://integrate.api.nvidia.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer nvapi-YOUR_API_KEY" \
  -d '{
    "model": "meta/llama-3.1-70b-instruct",
    "messages": [
      {
        "role": "system",
        "content": "You are a helpful AI assistant."
      },
      {
        "role": "user",
        "content": "你好，请介绍一下 NVIDIA NIM。"
      }
    ],
    "max_tokens": 1024,
    "temperature": 0.7
  }'

curl https://integrate.api.nvidia.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer nvapi-YOUR_API_KEY" \
  -d '{
    "model": "meta/llama-3.1-70b-instruct",
    "messages": [{"role": "user", "content": "你好"}],
    "stream": true
  }'

npm install openai

import OpenAI from 'openai';

const client = new OpenAI({
  baseURL: 'https://integrate.api.nvidia.com/v1',
  apiKey: 'nvapi-YOUR_API_KEY',
});

async function main() {
  const completion = await client.chat.completions.create({
    model: 'meta/llama-3.1-70b-instruct',
    messages: [
      { role: 'system', content: 'You are a helpful AI assistant.' },
      { role: 'user', content: '介绍一下 NVIDIA GPU 的优势。' }
    ],
    max_tokens: 1024,
    temperature: 0.7,
  });

  console.log(completion.choices[0].message.content);
  console.log(`\n使用 Tokens: ${completion.usage.total_tokens}`);
}

main();

from openai import OpenAI

client = OpenAI(
    base_url="https://integrate.api.nvidia.com/v1",
    api_key="nvapi-YOUR_API_KEY"
)

def customer_service_bot(user_question, context=""):
    """智能客服助手"""
    system_prompt = f"""你是一个专业的客服助手。
请基于以下知识库回答用户问题：

{context}

回答要求：
- 专业、友好、简洁
- 如果不确定，建议联系人工客服
- 提供具体的解决方案
"""
    
    response = client.chat.completions.create(
        model="meta/llama-3.1-70b-instruct",
        messages=[
            {"role": "system", "content": system_prompt},
            {"role": "user", "content": user_question}
        ],
        temperature=0.3  # 降低温度以获得更确定的回答
    )
    
    return response.choices[0].message.content

# 使用示例
knowledge_base = """
产品信息：
- 支持 7x24 小时在线客服
- 提供 30 天无理由退换货
- 全国包邮，偏远地区除外
"""

answer = customer_service_bot("如何退货？", knowledge_base)
print(answer)

from openai import OpenAI

client = OpenAI(
    base_url="https://integrate.api.nvidia.com/v1",
    api_key="nvapi-YOUR_API_KEY"
)

def code_review(code, language="Python"):
    """代码审查助手"""
    response = client.chat.completions.create(
        model="meta/llama-3.1-70b-instruct",
        messages=[
            {
                "role": "system",
                "content": f"你是一个专业的 {language} 代码审查专家。"
                           "请从以下方面审查代码：\n"
                           "1. 代码质量和可读性\n"
                           "2. 潜在的 bug 和安全问题\n"
                           "3. 性能优化建议\n"
                           "4. 最佳实践建议"
            },
            {
                "role": "user",
                "content": f"请审查以下代码：\n\n```{language.lower()}\n{code}\n```"
            }
        ],
        temperature=0.2
    )
    
    return response.choices[0].message.content

# 使用示例
code_to_review = """
def calculate_sum(numbers):
    sum = 0
    for i in range(len(numbers)):
        sum = sum + numbers[i]
    return sum
"""

review_result = code_review(code_to_review)
print(review_result)

from openai import OpenAI
import base64

client = OpenAI(
    base_url="https://integrate.api.nvidia.com/v1",
    api_key="nvapi-YOUR_API_KEY"
)

def analyze_image(image_url, question="描述这张图片"):
    """图像分析"""
    response = client.chat.completions.create(
        model="meta/llama-3.2-90b-vision-instruct",
        messages=[
            {
                "role": "user",
                "content": [
                    {"type": "text", "text": question},
                    {
                        "type": "image_url",
                        "image_url": {"url": image_url}
                    }
                ]
            }
        ]
    )
    
    return response.choices[0].message.content

# 使用示例
result = analyze_image(
    "https://example.com/product.jpg",
    "这个产品有什么特点？"
)
print(result)