Update README.md

README.md

@@ -40,3 +40,28 @@ merge_method: linear
 dtype: float16
 
 ```
+
+``` python
+
+
+import transformers
+import torch
+
+from transformers import AutoTokenizer, MixtralForCausalLM
+device = "cuda" # the device to load the model onto
+
+model = "{{ username }}/{{ model_name }}"
+imodel = MixtralForCausalLM.from_pretrained(model)
+tokenizer = AutoTokenizer.from_pretrained(model)
+
+
+inputs = tokenizer(prompt, return_tensors="pt")
+
+# Generate
+generate_ids = imodel.generate(inputs.input_ids, max_length=30)
+tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
+
+
+
+```
+