Update app.py

app.py

@@ -98,12 +98,12 @@ def load_image(image_file, input_size=448, max_num=12):
     return pixel_values
 
 model = AutoModel.from_pretrained(
-    "5CD-AI/Vintern-4B-v1",
+    "5CD-AI/Vintern-3B-beta",
     torch_dtype=torch.bfloat16,
     low_cpu_mem_usage=True,
     trust_remote_code=True,
 ).eval().cuda()
-tokenizer = AutoTokenizer.from_pretrained("5CD-AI/Vintern-4B-v1", trust_remote_code=True, use_fast=False)
+tokenizer = AutoTokenizer.from_pretrained("5CD-AI/Vintern-3B-beta", trust_remote_code=True, use_fast=False)
 
 
 @spaces.GPU
@@ -203,11 +203,11 @@ button.svelte-1lcyrx4[aria-label="user's message: a file of type image/jpeg, "]
 
 demo = gr.ChatInterface(
     fn=chat,
-    description="""Try [Vintern-4B-v1](https://huggingface.co/5CD-AI/Vintern-4B-v1) in this demo. Vintern-4B-v1 consists of [InternViT-300M-448px](https://huggingface.co/OpenGVLab/InternViT-300M-448px), an MLP projector, and [Sailor-4B-Chat](https://huggingface.co/sail/Sailor-4B-Chat).""",
+    description="""Try [Vintern-4B-v1](https://huggingface.co/5CD-AI/Vintern-3B-beta) in this demo. Vintern-3B-beta consists of [InternViT-300M-448px](https://huggingface.co/OpenGVLab/InternViT-300M-448px), an MLP projector, and [Qwen2.5-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct).""",
     examples=[{"text": "Mô tả hình ảnh.", "files":["./demo_3.jpg"]},
               {"text": "Trích xuất các thông tin từ ảnh.", "files":["./demo_1.jpg"]}, 
               {"text": "Mô tả hình ảnh một cách chi tiết.", "files":["./demo_2.jpg"]}],
-    title="❄️ Vintern-4B-v1 ❄️",
+    title="❄️ Vintern-3B-beta ❄️",
     multimodal=True,
     css=CSS
 )