Update app.py

app.py

@@ -1,16 +1,22 @@
 import gradio as gr
-from transformers import AutoTokenizer, AutoModelForCausalLM
+from transformers import AutoTokenizer, AutoModelForCausalLM, AutoConfig
 import torch
 
 print("Loading model...")
 MODEL_ID = "newtechdevng/qwen-math-tutor"
+
 tokenizer = AutoTokenizer.from_pretrained(MODEL_ID)
+
+# Load config and strip quantization settings
+config = AutoConfig.from_pretrained(MODEL_ID)
+config.quantization_config = {}  # ← remove baked-in quantization
+
 model = AutoModelForCausalLM.from_pretrained(
     MODEL_ID,
-    torch_dtype           = torch.float32,
-    device_map            = "cpu",
-    quantization_config   = None,
+    config  = config,
+    dtype   = torch.float32,  # ← fixed from torch_dtype
 )
+model = model.to("cpu")
 print("Model loaded!")
 
 def solve(question, history):
@@ -40,7 +46,7 @@ def solve(question, history):
         skip_special_tokens=True
     )
 
-with gr.Blocks(title="Math Tutor AI") as demo:  # ← theme removed from here
+with gr.Blocks(title="Math Tutor AI") as demo:
     gr.Markdown("""
     # 🧮 Math Tutor AI
     ### Powered by Qwen2.5-Math — Fine-tuned for NCERT & competitive math
@@ -74,4 +80,4 @@ with gr.Blocks(title="Math Tutor AI") as demo:  # ← theme removed from here
     submit.click(respond, [question, chatbot], [question, chatbot])
     question.submit(respond, [question, chatbot], [question, chatbot])
 
-demo.launch(theme=gr.themes.Soft())  # ← theme moved to launch()
+demo.launch(theme=gr.themes.Soft())