Create app.py

app.py

@@ -0,0 +1,102 @@
+import gradio as gr
+from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
+import torch
+
+# Model loading with optimized settings
+MODEL_NAME = "Qwen/Qwen3-0.6B"
+cache_dir = "./model_cache"
+
+tokenizer = AutoTokenizer.from_pretrained(
+    MODEL_NAME,
+    trust_remote_code=True,
+    cache_dir=cache_dir
+)
+
+model = AutoModelForCausalLM.from_pretrained(
+    MODEL_NAME,
+    trust_remote_code=True,
+    torch_dtype=torch.float16,
+    device_map="auto",
+    cache_dir=cache_dir
+).eval()
+
+# Create text generation pipeline
+text_generator = pipeline(
+    "text-generation",
+    model=model,
+    tokenizer=tokenizer,
+    device=0 if torch.cuda.is_available() else -1,
+    pad_token_id=tokenizer.eos_token_id
+)
+
+def generate_response(prompt, max_new_tokens=256, temperature=0.7, top_p=0.9):
+    """Generate response with safe defaults"""
+    try:
+        response = text_generator(
+            prompt,
+            max_new_tokens=int(max_new_tokens),
+            temperature=float(temperature),
+            top_p=float(top_p),
+            do_sample=True,
+            truncation=True
+        )
+        return response[0]["generated_text"]
+    except Exception as e:
+        return f"Error: {str(e)}"
+
+# Gradio interface with advanced settings
+with gr.Blocks(theme="soft", title="Qwen3-0.6B Chat Interface") as demo:
+    gr.Markdown("# 🧠 Qwen3-0.6B Text-to-Text Chat")
+    gr.Markdown("Powered by HuggingFace Transformers and Gradio")
+    
+    with gr.Row():
+        with gr.Column():
+            prompt = gr.Textbox(
+                label="User Input",
+                placeholder="Ask me anything...",
+                lines=5
+            )
+            with gr.Accordion("Advanced Settings", open=False):
+                max_new_tokens = gr.Slider(
+                    minimum=32, 
+                    maximum=512, 
+                    value=256, 
+                    step=32,
+                    label="Max New Tokens"
+                )
+                temperature = gr.Slider(
+                    minimum=0.1, 
+                    maximum=1.0, 
+                    value=0.7, 
+                    step=0.1,
+                    label="Temperature"
+                )
+                top_p = gr.Slider(
+                    minimum=0.1, 
+                    maximum=1.0, 
+                    value=0.9, 
+                    step=0.1,
+                    label="Top-p Sampling"
+                )
+            
+        with gr.Column():
+            output = gr.Textbox(label="Model Response", lines=10)
+    
+    submit = gr.Button("💬 Generate Response")
+    submit.click(
+        fn=generate_response,
+        inputs=[prompt, max_new_tokens, temperature, top_p],
+        outputs=output
+    )
+    
+    gr.Examples(
+        examples=[
+            ["Explain quantum computing in simple terms"],
+            ["Write a poem about autumn leaves"],
+            ["Solve this math problem: 2x + 5 = 17"]
+        ],
+        inputs=prompt
+    )
+
+if __name__ == "__main__":
+    demo.launch()