Update README.md

README.md

@@ -54,6 +54,24 @@ decoded_output = tokenizer.decode(outputs[0], skip_special_tokens=True)
 print(decoded_output)
 ```
 
+LOADING THE MODEL DIRCTLY: 
+
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+import torch
+
+# Use the merged model name from your Hugging Face repository
+model_name = "vishal042002/Llama3.2-3b-Instruct-ClinicalSurgery"
+model = AutoModelForCausalLM.from_pretrained(
+    model_name,
+    torch_dtype=torch.float16,
+    device_map="auto"
+)
+
+# Load the tokenizer
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+```
+
 
 This model is designed to: