Update app.py

app.py

@@ -7,8 +7,9 @@ from huggingface_hub import hf_hub_download
 
 llm = Llama(
     model_path=hf_hub_download(
-        repo_id=os.environ.get("REPO_ID", "microsoft/Phi-3-mini-4k-instruct-gguf"),
-        filename=os.environ.get("MODEL_FILE", "Phi-3-mini-4k-instruct-q4.gguf"),
+        repo_id=os.environ.get("REPO_ID", "unsloth/NVIDIA-Nemotron-3-Nano-4B-GGUF:Q4_K_M"),# Run inference directly in the terminal:
+
+        filename=os.environ.get("MODEL_FILE", "NVIDIA-Nemotron-3-Nano-4B-GGUF:Q4_K_M"),
     ),
     n_ctx=2048,
     n_gpu_layers=50, # change n_gpu_layers if you have more or less VRAM