Update app.py

app.py

@@ -12,17 +12,17 @@ model = AutoModelForCausalLM.from_pretrained(MODEL_ID, torch_dtype=torch.bfloat1
 # -------------------------------------------------
 # Optional tool(s)
 # -------------------------------------------------
-TOOLS = [{
-    "name": "get_weather",
-    "description": "Get the current weather in a given city",
-    "parameters": {
-        "type": "object",
-        "properties": {
-            "city": {"type": "string", "description": "City name"}
-        },
-        "required": ["city"]
-    }
-}]
+# TOOLS = [{
+#     "name": "get_weather",
+#     "description": "Get the current weather in a given city",
+#     "parameters": {
+#         "type": "object",
+#         "properties": {
+#             "city": {"type": "string", "description": "City name"}
+#         },
+#         "required": ["city"]
+#     }
+# }]
 
 # -------------------------------------------------
 # Helpers
@@ -37,14 +37,14 @@ def build_messages(history, enable_thinking: bool):
     messages.insert(0, {"role": "system", "content": system_flag})
     return messages
 
-def chat_fn(history, enable_thinking, temperature, top_p):
+def chat_fn(history, enable_thinking, temperature, top_p, top_k, repetition_penalty):
     """Generate a streaming response."""
     messages = build_messages(history, enable_thinking)
     text = tokenizer.apply_chat_template(
         messages,
         tokenize=False,
         add_generation_prompt=True,
-        xml_tools=TOOLS
+        # xml_tools=TOOLS
     )
     inputs = tokenizer(text, return_tensors="pt").to(DEVICE)
 
@@ -54,8 +54,8 @@ def chat_fn(history, enable_thinking, temperature, top_p):
         do_sample=True,
         temperature=temperature,
         top_p=top_p,
-        top_k=20,
-        repetition_penalty=1.1,
+        top_k=top_k,
+        repetition_penalty=repetition_penalty,
         pad_token_id=tokenizer.eos_token_id,
         streamer=None          # we'll yield manually
     )
@@ -77,6 +77,8 @@ with gr.Blocks(title="SmolLM3-3B Chat") as demo:
         enable_think = gr.Checkbox(label="Enable Extended Thinking (/think)", value=False)
         temperature = gr.Slider(0.0, 1.0, value=0.6, label="Temperature")
         top_p = gr.Slider(0.0, 1.0, value=0.95, label="Top-p")
+        top_k = gr.Slider(1,40,value=20,label="Top_k")
+        repetition_penalty = gr.Slider(1.0,1.4,value=1.1,label="Repetition_Penalty")
     chatbot = gr.Chatbot(type="messages")
     msg = gr.Textbox(placeholder="Type your message here…", lines=1)
     clear = gr.Button("Clear")
@@ -87,7 +89,7 @@ with gr.Blocks(title="SmolLM3-3B Chat") as demo:
     msg.submit(
         user_fn, [msg, chatbot], [msg, chatbot], queue=False
     ).then(
-        chat_fn, [chatbot, enable_think, temperature, top_p], chatbot
+        chat_fn, [chatbot, enable_think, temperature, top_p, top_k, repetition_penalty], chatbot
     )
     clear.click(lambda: None, None, chatbot, queue=False)