update app

app.py

@@ -17,9 +17,10 @@ model = AutoModelForCausalLM.from_pretrained(
     device_map="cpu",
     trust_remote_code=True,
 )
-if not tokenizer.chat_template:
-    tokenizer.chat_template = """{% for message in messages %}{{ message['role'] }}: {{ message['content'] }}
-{% endfor %}Assistant:"""
+model.config.use_cache = True
+# if not tokenizer.chat_template:
+#     tokenizer.chat_template = """{% for message in messages %}{{ message['role'] }}: {{ message['content'] }}
+# {% endfor %}Assistant:"""
 
 # Request schema
 class ChatRequest(BaseModel):
@@ -29,19 +30,19 @@ class ChatRequest(BaseModel):
 @app.post("/chat")
 def chat(req: ChatRequest):
     # Format input sesuai template Qwen
-    # text = tokenizer.apply_chat_template(
-    #     req.messages,
-    #     tokenize=False,
-    #     add_generation_prompt=True
-    # )
-    # inputs = tokenizer(text, return_tensors="pt").to(model.device)
+    text = tokenizer.apply_chat_template(
+        req.messages,
+        tokenize=False,
+        add_generation_prompt=True
+    )
+    inputs = tokenizer(text, return_tensors="pt").to(model.device)
 
-    prompt = ""
-    for msg in req.messages:
-        prompt += f"{msg['role']}: {msg['content']}\n"
-    prompt += "assistant:"
+    # prompt = ""
+    # for msg in req.messages:
+    #     prompt += f"{msg['role']}: {msg['content']}\n"
+    # prompt += "assistant:"
 
-    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
+    # inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
 
     # Generate
     outputs = model.generate(