Update README.md

README.md

@@ -51,7 +51,19 @@ TinyMistral-6x248M is a Mixure of Experts (MoE) made with the following models u
 * [Felladrin/TinyMistral-248M-SFT-v4](https://huggingface.co/Felladrin/TinyMistral-248M-SFT-v4)
 * [Locutusque/TinyMistral-248M-v2-Instruct](https://huggingface.co/Locutusque/TinyMistral-248M-v2-Instruct)
 
-The resulting model is then pre-trained on 600,000 examples of nampdn-ai/mini-peS2o
+The resulting model is then pre-trained on 600,000 examples of nampdn-ai/mini-peS2o.
+
+We don't recommend using the Inference API as the model has serious performance degradation.
+
+### Recommended inference parameters
+
+```
+do_sample: true
+temperature: 0.2
+top_p: 0.14
+top_k: 12
+repetition_penalty: 1.15
+```
 
 ## 🧩 Configuration