update sth. at 2025-01-05 00:57:05

Dockerfile

@@ -17,4 +17,4 @@ RUN pip install --no-cache-dir -r requirements.txt
 EXPOSE 7860
 
 # 运行 Flask 应用
-CMD ["python", "app241224.py"]
+CMD ["python", "app250105.py"]

app241224.py → app250105.py

@@ -16,9 +16,8 @@ def get_models():
         "object": "list",
         "data": [
             {"id": "Qwen2.5-72B", "object": "model", "created": 0, "owned_by": "Qwen"},
-            {"id": "Llama-3.1-Nemotron-70B", "object": "model", "created": 0, "owned_by": "Nemotron"},
-            {"id": "NVLM-D-72B", "object": "model", "created": 0, "owned_by": "NVDIA"},
-            {"id": "DeepSeek-Coder-V2", "object": "model", "created": 0, "owned_by": "DeepSeek"},
+            {"id": "Llama3.3-70B", "object": "model", "created": 0, "owned_by": "Nemotron"},
+            {"id": "Pixtral-124B", "object": "model", "created": 0, "owned_by": "Pixtral"},
             {"id": "Qwen2.5-Coder-32B", "object": "model", "created": 0, "owned_by": "Qwen"},
         ]
     }

runway.py

@@ -0,0 +1,34 @@
+import requests
+
+# 设置 API 密钥和请求的 URL
+url = 'https://sanbo1200-degpt.hf.space/hf/v1/chat/completions'
+
+# 构建请求头
+headers = {
+    # 'Authorization': f'Bearer {api_key}',
+    'Content-Type': 'application/json',
+}
+
+# 构建请求体
+data = {
+    'model': 'gpt-4o',  # 或者使用其他可用的模型
+    'messages': [
+        {'role': 'user', 'content': '你好，你是谁？'},
+    ],
+    'max_tokens': 100,  # 设置生成的最大 token 数量
+}
+
+# 发送 POST 请求
+response = requests.post(url, headers=headers, json=data)
+
+# 检查响应状态
+if response.status_code == 200:
+    # print('请求成功',response.text)
+    # 解析 JSON 响应
+    response_data = response.json()
+    # 获取生成的文本
+    generated_text = response_data['choices'][0]['message']['content']
+    print('ChatGPT 的回复:', generated_text)
+else:
+    print('请求失败，状态码:', response.status_code)
+    print('错误信息:', response.text)