import openai
import time
import requests
import json


client = openai.Client(
    api_key="cannot be empty",
    base_url="http://127.0.0.1:9997/v1"
)

# 记录开始时间
start_time = time.time()

a = client.embeddings.create(
    model='bge-m3',
    input=["What is the capital of China?"]
)

# 记录结束时间
end_time = time.time()

#print(a)
print(f"\n耗时: {end_time - start_time:.4f} 秒")

# 记录开始时间
start_time = time.time()

a = client.embeddings.create(
    model='Qwen3-Embedding-0.6B',
    input=["What is the capital of China?"]
)

# 记录结束时间
end_time = time.time()

#print(a)
print(f"\n耗时: {end_time - start_time:.4f} 秒")

# ========== HTTP API 测试 ==========
print("\n" + "="*50)
print("HTTP API 测试")
print("="*50)

# 配置
XINFERENCE_HOST = "127.0.0.1"
XINFERENCE_PORT = "9997"
base_url = f"http://{XINFERENCE_HOST}:{XINFERENCE_PORT}/v1/embeddings"

# 测试 bge-m3 模型
print("\n测试模型: bge-m3")
start_time = time.time()

response = requests.post(
    base_url,
    headers={
        'accept': 'application/json',
        'Content-Type': 'application/json'
    },
    json={
        "model": "bge-m3",
        "input": "What is the capital of China?"
    }
)

end_time = time.time()

if response.status_code == 200:
    result = response.json()
    print(f"状态码: {response.status_code}")
    print(f"模型: {result.get('model', 'N/A')}")
    print(f"使用token数: {result.get('usage', {}).get('total_tokens', 'N/A')}")
    print(f"嵌入向量维度: {len(result.get('data', [{}])[0].get('embedding', []))}")
    print(f"耗时: {end_time - start_time:.4f} 秒")
else:
    print(f"请求失败，状态码: {response.status_code}")
    print(f"错误信息: {response.text}")

# 测试 Qwen3-Embedding-0.6B 模型
print("\n测试模型: Qwen3-Embedding-0.6B")
start_time = time.time()

response = requests.post(
    base_url,
    headers={
        'accept': 'application/json',
        'Content-Type': 'application/json'
    },
    json={
        "model": "Qwen3-Embedding-0.6B",
        "input": "What is the capital of China?"
    }
)

end_time = time.time()

if response.status_code == 200:
    result = response.json()
    print(f"状态码: {response.status_code}")
    print(f"模型: {result.get('model', 'N/A')}")
    print(f"使用token数: {result.get('usage', {}).get('total_tokens', 'N/A')}")
    print(f"嵌入向量维度: {len(result.get('data', [{}])[0].get('embedding', []))}")
    print(f"耗时: {end_time - start_time:.4f} 秒")
else:
    print(f"请求失败，状态码: {response.status_code}")
    print(f"错误信息: {response.text}")