Create app.py

app.py

@@ -0,0 +1,40 @@
+import gradio as gr
+from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
+
+# Replace "username/llama3.3" with your actual model repository
+MODEL_NAME = "meta-llama/Llama-2-7b-chat-hf"
+
+# Load tokenizer and model
+tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
+model = AutoModelForCausalLM.from_pretrained(MODEL_NAME, torch_dtype="auto")
+
+# Create a text-generation pipeline
+text_gen = pipeline(
+    "text-generation",
+    model=model,
+    tokenizer=tokenizer,
+    max_length=512,
+    do_sample=True,
+    temperature=0.7
+)
+
+def chat(user_input):
+    """
+    Simple chat function that prepends user input to a system prompt (if needed)
+    and returns the model's text generation.
+    """
+    # If you have a special prompt format for a chat model, add it here.
+    # For a generic chat, you can just send the user_input:
+    outputs = text_gen(user_input, max_length=512)
+    return outputs[0]["generated_text"]
+
+demo = gr.Interface(
+    fn=chat,
+    inputs="text",
+    outputs="text",
+    title="LLaMA3.3 Chat (Example)",
+    description="A chat interface for the LLaMA-based model named 'llama3.3'."
+)
+
+if __name__ == "__main__":
+    demo.launch()