Update app.py

app.py

@@ -1,23 +1,16 @@
 import gradio as gr
 from llama_cpp import Llama
 
-# Link to the model in another Hugging Face repository
-repo_name = "google/gemma-7b-it-GGUF"
-model_path = "gemma-7b-it-gguf"  # Ensure this matches the actual file name in the repo
-model = Llama(hf_repo=repo_name, hf_file=model_path, n_ctx=512)  # No need to worry about downloading manually
+model = Llama.from_pretrained(repo_id="google/gemma-7b-it-GGUF", filename="gemma-7b-it.gguf")
 
-# Define the chatbot function
 def chat(input_text):
-    output = model(input_text, max_tokens=150, temperature=0.7, top_p=0.9)
+    output = model(f"<bos><start_of_turn>user\n{input_text}<end_of_turn>\n<start_of_turn>model\n", stop=["<start_of_turn>model", "\n"])
     return output["choices"][0]["text"]
 
-# Create the Gradio interface
 interface = gr.Interface(
     fn=chat,
     inputs="text",
-    outputs="text",
-    title="Gemma 7B Instruct GGUF Chatbot allowed for Space Use by @mikemin027"
+    outputs="text"
 )
 
-# Launch the interface
-interface.launch()
+interface.launch()