Update app.py

app.py

@@ -17,8 +17,9 @@ model_id = "OpenGVLab/InternVL2_5-2B"
 model = AutoModel.from_pretrained(
     model_id, 
     trust_remote_code=True, 
-    torch_dtype=torch.bfloat16,  # <--- CRITICAL CHANGE: Reduces RAM usage by 50%
-    low_cpu_mem_usage=True       # <--- HELPFUL: Prevents RAM spikes during loading
+    torch_dtype=torch.bfloat16,  
+    low_cpu_mem_usage=True,
+    revision="65b9340"
 ).to(device).eval()
 tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)