Add Space files

app.py

@@ -0,0 +1,95 @@
+"""
+Stack 2.9 - HuggingFace Space
+Code Assistant using Qwen2.5-Coder
+"""
+import gradio as gr
+from transformers import AutoModelForCausalLM, AutoTokenizer
+import torch
+
+# Load base model - uses 1.5B model which fits in free tier
+MODEL_NAME = "Qwen/Qwen2.5-Coder-1.5B"
+
+print(f"Loading {MODEL_NAME}...")
+tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME, trust_remote_code=True)
+model = AutoModelForCausalLM.from_pretrained(
+    MODEL_NAME,
+    torch_dtype=torch.float16,
+    device_map="auto",
+    trust_remote_code=True
+)
+print("Model loaded!")
+
+def generate(prompt, system_prompt="You are a helpful coding assistant.", max_tokens=512, temperature=0.7):
+    """Generate response from the model"""
+    messages = [
+        {"role": "system", "content": system_prompt},
+        {"role": "user", "content": prompt}
+    ]
+
+    text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
+    inputs = tokenizer([text], return_tensors="pt").to(model.device)
+
+    outputs = model.generate(
+        **inputs,
+        max_new_tokens=max_tokens,
+        temperature=temperature,
+        do_sample=True,
+        pad_token_id=tokenizer.pad_token_id
+    )
+
+    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    # Remove the input prompt from response
+    return response[len(text):].strip()
+
+# Build Gradio UI
+with gr.Blocks(title="Stack 2.9 - Code Assistant") as demo:
+    gr.Markdown("""
+    # Stack 2.9 - Code Assistant
+    **Powered by Qwen2.5-Coder-1.5B** fine-tuned on Stack Overflow data
+
+    Write code, debug, or ask programming questions!
+    """)
+
+    with gr.Row():
+        with gr.Column(scale=1):
+            system_prompt = gr.Textbox(
+                label="System Prompt",
+                value="You are a helpful coding assistant specialized in programming.",
+                lines=3
+            )
+            prompt = gr.Textbox(
+                label="Your Message",
+                placeholder="Write a Python function to calculate fibonacci numbers...",
+                lines=6
+            )
+            with gr.Row():
+                max_tokens = gr.Slider(32, 1024, value=512, step=32, label="Max Tokens")
+                temperature = gr.Slider(0.1, 1.5, value=0.7, step=0.1, label="Temperature")
+            submit = gr.Button("Generate", variant="primary")
+
+        with gr.Column(scale=2):
+            output = gr.Textbox(label="Response", lines=15)
+
+    examples = [
+        ["Write a Python function to calculate fibonacci numbers"],
+        ["Explain what this code does: def foo(x): return x * 2"],
+        ["Debug this code: for i in range(10): print(i)"],
+        ["Write a SQL query to find duplicate emails"],
+    ]
+
+    gr.Examples(examples=examples, inputs=[prompt])
+
+    submit.click(
+        fn=generate,
+        inputs=[prompt, system_prompt, max_tokens, temperature],
+        outputs=output
+    )
+
+    prompt.submit(
+        fn=generate,
+        inputs=[prompt, system_prompt, max_tokens, temperature],
+        outputs=output
+    )
+
+if __name__ == "__main__":
+    demo.launch(server_name="0.0.0.0", server_port=7860)