Update README.md

README.md

@@ -1,6 +1,12 @@
 ---
 license: apache-2.0
 ---
+# About the model
+Predibase-T2T-32B-RFT is a 32-billion parameter transformer model fine-tuned with Reinforcement Fine-Tuning (RFT) on the Predibase platform. It is optimized for high-quality text generation with strong generalization across reasoning, instruction-following, and domain-specific tasks.
+
+Unlike traditional supervised fine-tuning approaches, Predibase-T2T-32B-RFT leverages RFT to adapt model behavior interactively, optimizing for downstream task quality with minimal labeled data. This makes it a highly cost-efficient and performant alternative to proprietary LLMs.
+
+The model is fine-tuned on a diverse set of reward functions, allowing it to dynamically adjust responses based on contextual understanding.
 
 # How to use this model