Update README.md

README.md

@@ -193,7 +193,7 @@ For best results, structure your prompts to align with the model's internal oper
 
 *   **Not a Generalist:** This model has been intensely specialized. Its performance on tasks outside its core training (e.g., highly technical code generation, casual chit-chat) may not be state-of-the-art.
 *   **"Overly" Sophisticated:** Due to its training on elite exemplars, its default communication style is professional and highly structured. It may sometimes sound "too formal" for very casual use cases.
-*   **Inherited Biases:** As a derivative of Llama 3.1, it may inherit biases present in the original pre-training data.
+*   **Inherited Biases:** As a derivative of Qwen3-8B, it may inherit biases present in the original pre-training data.
 
 ### **Training Procedure**