Switch to TinyLlama model to fit memory constraints

main.py

@@ -10,7 +10,7 @@ import json
 # 1. Configuration
 # ----------------------------
 
-MODEL_NAME = "deepseek-ai/deepseek-coder-6.7b-instruct"
+MODEL_NAME = "TinyLlama/TinyLlama-1.1B-Chat-v1.0"
 # The device will be automatically handled by device_map="auto"
 # but we can keep this for logging.
 DEVICE = "cuda" if torch.cuda.is_available() else "cpu"