Update app.py

app.py

@@ -7,8 +7,8 @@ from huggingface_hub import hf_hub_download
 # Initialize Llama model from Hugging Face
 llm = Llama(
     model_path=hf_hub_download(
-        repo_id=os.environ.get("REPO_ID", "mradermacher/Atlas-Chat-9B-GGUF"),
-        filename=os.environ.get("MODEL_FILE", "Atlas-Chat-9B.Q4_K_M.gguf"),
+        repo_id=os.environ.get("REPO_ID", "mradermacher/Atlas-Chat-2B-GGUF"),
+        filename=os.environ.get("MODEL_FILE", "Atlas-Chat-2B.Q8_0.gguf"),
     ),
     n_ctx=4096,
     n_gpu_layers=-1,
@@ -65,7 +65,7 @@ def generate_text(
 # Define the Gradio interface
 demo = gr.ChatInterface(
     generate_text,
-    title="Llama-3.1-8B-Instruct-Reasoner",
+    title="using Atlas-Chat-2B | I had to switch to the 2B model because the 9B was too much for this space!",
     description="Running LLM with https://github.com/abetlen/llama-cpp-python",
     examples=[
         ['How to setup a human base on Mars? Give short answer.'],