Create app.py

app.py

@@ -0,0 +1,41 @@
+import torch
+from transformers import AutoTokenizer, AutoModelForCausalLM
+import gradio as gr
+
+# 修改为你上传的模型文件夹路径
+model_path = "./my-deepseek"
+
+# 加载模型和分词器
+tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True)
+model = AutoModelForCausalLM.from_pretrained(
+    model_path, 
+    trust_remote_code=True, 
+    torch_dtype=torch.float32  # 若使用 GPU，可改为 torch.float16
+)
+model.eval()
+
+def chat(prompt, max_new_tokens=512):
+    inputs = tokenizer(prompt, return_tensors="pt")
+    with torch.no_grad():
+        outputs = model.generate(
+            **inputs,
+            max_new_tokens=max_new_tokens,
+            do_sample=True,
+            top_p=0.95,
+            temperature=0.8,
+            pad_token_id=tokenizer.eos_token_id,
+        )
+    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    return response[len(prompt):]  # 返回去除原始prompt后的回答部分
+
+# 创建 Gradio 界面
+gr.Interface(
+    fn=chat,
+    inputs=[
+        gr.Textbox(label="你的提问", lines=3, placeholder="请输入 prompt..."),
+        gr.Slider(128, 1024, step=64, value=512, label="最大生成长度")
+    ],
+    outputs="text",
+    title="🧠 DeepSeek-R1 Chat Demo",
+    description="使用你本地上传的 DeepSeek-R1 模型运行的聊天机器人。"
+).launch()