Create app.py

app.py

@@ -0,0 +1,42 @@
+import gradio as gr
+from transformers import AutoTokenizer, AutoModelForCausalLM
+from peft import PeftModel
+
+# Model path
+model_name = "Qwen/Qwen3-0.6B"
+peft_model_path = "gauravprasadgp/Qwen3-0.6B_demeaner_hinglish"
+
+# Load tokenizer and model
+tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
+base_model = AutoModelForCausalLM.from_pretrained(model_name, trust_remote_code=True, device_map="auto")
+model = PeftModel.from_pretrained(base_model, peft_model_path)
+
+def generate_response(prompt, max_tokens, temperature, top_p):
+    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
+    outputs = model.generate(
+        **inputs,
+        max_new_tokens=max_tokens,
+        temperature=temperature,
+        top_p=top_p,
+        do_sample=True
+    )
+    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    return response[len(prompt):].strip()  # Trim input from output
+
+# UI
+iface = gr.Interface(
+    fn=generate_response,
+    inputs=[
+        gr.Textbox(label="Did Gaurav asked you to try me ???", lines=4, placeholder="Type something..."),
+        gr.Slider(20, 512, value=128, step=8, label="Max Tokens"),
+        gr.Slider(0.1, 1.0, value=0.7, step=0.1, label="Temperature"),
+        gr.Slider(0.1, 1.0, value=0.9, step=0.1, label="Top-p (nucleus sampling)")
+    ],
+    outputs=gr.Textbox(label="Model Response"),
+    title="Hinglish Demeaner Model (Qwen3-0.6B)",
+    description="This is a fine-tuned Qwen model to generate Hinglish responses with stylistic modifications.",
+    theme="default"
+)
+
+if __name__ == "__main__":
+    iface.launch()