Update app.py

app.py

@@ -1,5 +1,5 @@
 import gradio as gr
-from transformers import AutoProcessor, AutoModelForVision2Seq, TextIteratorStreamer
+from transformers import AutoProcessor, AutoModel, TextIteratorStreamer
 from threading import Thread
 import re
 import time
@@ -10,8 +10,12 @@ import spaces
 #subprocess.run('pip install flash-attn --no-build-isolation', env={'FLASH_ATTENTION_SKIP_CUDA_BUILD': "TRUE"}, shell=True)
 
 
-processor = AutoProcessor.from_pretrained("HuggingFaceTB/SmolVLM-Instruct")
-model = AutoModelForVision2Seq.from_pretrained("HuggingFaceTB/SmolVLM-Instruct", 
+processor = AutoProcessor.from_pretrained(
+    "rhymes-ai/Aria",
+    revision="0e2a8319",
+)
+model = AutoModel.from_pretrained("rhymes-ai/Aria", 
+        revision="f7c92bba",
         torch_dtype=torch.bfloat16,
         #_attn_implementation="flash_attention_2"
         ).to("cuda")