Update rag_hf.py

rag_hf.py

@@ -14,7 +14,8 @@ from dotenv import load_dotenv
 # === CONFIGURATION ===
 load_dotenv()
 
-ENDPOINT_URL = os.getenv("HF_ENDPOINT")
+ENDPOINT_URL = "https://api-inference.huggingface.co/models/mistralai/Mistral-7B-Instruct-v0.3"
+
 HF_API_TOKEN = os.getenv("HF_API_TOKEN")
 EMBEDDING_MODEL = "intfloat/multilingual-e5-base"
 DEVICE = "cuda" if torch.cuda.is_available() else "cpu"