bf16

app.py

@@ -2,12 +2,13 @@ import gradio as gr
 from unsloth import FastLanguageModel
 from transformers import TextIteratorStreamer
 from threading import Thread
+from torch import bfloat16
 
 # Load your fine-tuned model and tokenizer
 model, tokenizer = FastLanguageModel.from_pretrained(
     model_name=".",  # Path to your fine-tuned model
     max_seq_length=8192,
-    dtype='bf16',
+    dtype=bfloat16,
     load_in_4bit=False,
 )
 FastLanguageModel.for_inference(model)  # Enable optimized inference