Update README.md

README.md

@@ -28,7 +28,7 @@ The model is finetuned only on native telugu SFT data from above datasets and we
 ## Response: {response}
 ```
 
-# Usage with Unsloth
+# Inference with Unsloth
 
 ```python3
 from unsloth import FastLanguageModel
@@ -67,7 +67,7 @@ outputs = model.generate(**inputs, max_new_tokens = 300, use_cache = True)
 response = tokenizer.batch_decode(outputs)
 ```
 
-# Usage with HuggingFace
+# Inference with HuggingFace
 
 ```python3
 from peft import AutoPeftModelForCausalLM