Update README.md

README.md

@@ -11,6 +11,8 @@ inference: false
 
 # LLaMA 7B trained on the ru_turbo_alpaca, Russian instructions dataset
 
+**Important**: You should probably use [Saiga](https://huggingface.co/IlyaGusev/saiga_7b_lora). It has regular updates and should be better in every task.
+
 Based on [LLaMA 7B](https://huggingface.co/decapoda-research/llama-7b-hf).
 
 * This is an adapter-only version.