Skip to main content
Glama

TAPD Data Fetcher

test_api_compatibility.py3.81 kB
""" 测试API兼容性的脚本 测试SiliconFlow和DeepSeek两种API的调用 注意:现在默认使用SiliconFlow的'deepseek-ai/DeepSeek-V3.1'模型 """ import asyncio import aiohttp import sys import os # 添加项目根目录到Python路径 project_root = os.path.dirname(os.path.dirname(os.path.abspath(__file__))) sys.path.insert(0, project_root) from mcp_tools.common_utils import get_api_manager async def test_api_calls(): """测试不同API的调用""" api_manager = get_api_manager() test_prompt = "你好,请简单介绍一下自己。" async with aiohttp.ClientSession() as session: print("测试API兼容性...") print("=" * 50) # 测试SiliconFlow API(默认配置) print("\n🧪 测试1: SiliconFlow API 默认调用(默认使用deepseek-ai/DeepSeek-V3.1)") try: result = await api_manager.call_llm( prompt=test_prompt, session=session, max_tokens=100 ) print(f"✅ SiliconFlow API默认调用成功") print(f"📤 响应: {result[:200]}...") except Exception as e: print(f"❌ SiliconFlow API默认调用失败: {e}") print("=" * 50) # 测试SiliconFlow API(显式指定) print("\n🧪 测试2: SiliconFlow API 显式指定") try: result = await api_manager.call_llm( prompt=test_prompt, session=session, model="deepseek-ai/DeepSeek-V3.1", endpoint="https://api.siliconflow.cn/v1", max_tokens=100 ) print(f"✅ SiliconFlow API显式调用成功") print(f"📤 响应: {result[:200]}...") except Exception as e: print(f"❌ SiliconFlow API显式调用失败: {e}") print("=" * 50) # 测试DeepSeek API(需要显式指定endpoint) print("\n🧪 测试3: DeepSeek API 调用(显式指定endpoint)") try: result = await api_manager.call_llm( prompt=test_prompt, session=session, model="deepseek-chat", endpoint="https://api.deepseek.com/v1", max_tokens=100 ) print(f"✅ DeepSeek API调用成功") print(f"📤 响应: {result[:200]}...") except Exception as e: print(f"❌ DeepSeek API调用失败: {e}") print("=" * 50) # 测试DeepSeek Reasoner模型 print("\n🧪 测试4: DeepSeek Reasoner 模型") try: result = await api_manager.call_llm( prompt="请简单解释一下人工智能的基本概念。", session=session, model="deepseek-reasoner", endpoint="https://api.deepseek.com/v1", max_tokens=150 ) print(f"✅ DeepSeek Reasoner模型调用成功") print(f"📤 响应: {result[:200]}...") except Exception as e: print(f"❌ DeepSeek Reasoner模型调用失败: {e}") print("=" * 50) if __name__ == "__main__": print("🚀 开始API兼容性测试...") print("\n📋 环境变量检查:") print(f"SF_KEY: {'已设置' if os.getenv('SF_KEY') else '未设置'}") print(f"DS_KEY: {'已设置' if os.getenv('DS_KEY') else '未设置'}") print(f"SF_MODEL: {os.getenv('SF_MODEL', 'deepseek-ai/DeepSeek-V3.1 (默认)')}") print("\n💡 注意:现在默认使用SiliconFlow API的'deepseek-ai/DeepSeek-V3.1'模型") print(" 只有显式指定DeepSeek endpoint时才会调用DeepSeek API") asyncio.run(test_api_calls()) print("\n✅ 测试完成!")

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/OneCuriousLearner/MCPAgentRE'

If you have feedback or need assistance with the MCP directory API, please join our Discord server