Update app.py

app.py

@@ -1,19 +1,16 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
+from transformers import pipeline
 
 """
 For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
 """
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
-client = InferenceClient("Qwen/Qwen2.5-VL-3B-Instruct")
 
 # Use a pipeline as a high-level helper
-from transformers import pipeline
+pipe = pipeline("text-generation", model="vc3vc3/qwen3-0.6B-finetune")
 
 messages = [
-    {"role": "user", "content": "Who are you?"},
+    {"role": "user", "content": "Who are you? 用中文回答，风格调皮一些。"},
 ]
-pipe = pipeline("image-text-to-text", model="Qwen/Qwen2.5-VL-7B-Instruct")
 pipe(messages)
 
 def respond(
@@ -24,27 +21,28 @@ def respond(
     temperature,
     top_p,
 ):
-    messages = [{"role": "system", "content": system_message}]
-
+    # 拼接历史消息和当前消息为 prompt
+    prompt = system_message + "\n"
     for val in history:
         if val[0]:
-            messages.append({"role": "user", "content": val[0]})
+            prompt += f"用户: {val[0]}\n"
         if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
-
-    messages.append({"role": "user", "content": message})
+            prompt += f"助手: {val[1]}\n"
+    prompt += f"用户: {message}\n助手:"
 
+    # 使用 pipe 生成回复
     response = ""
-
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
+    for out in pipe(
+        prompt,
+        max_new_tokens=max_tokens,
         temperature=temperature,
         top_p=top_p,
+        do_sample=True,
+        return_full_text=False,
+        truncation=True,
+        stream=True,
     ):
-        token = message.choices[0].delta.content
-
+        token = out["generated_text"] if isinstance(out, dict) and "generated_text" in out else str(out)
         response += token
         yield response
 
@@ -70,4 +68,4 @@ demo = gr.ChatInterface(
 
 
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()