Update to Qwen2.5-VL model and clean up requirements

app.py

@@ -1,6 +1,6 @@
 import gradio as gr
 import spaces
-from transformers import Qwen2VLForConditionalGeneration, AutoProcessor
+from transformers import Qwen2VLForConditionalGeneration, AutoProcessor, Qwen2_5_VLForConditionalGeneration
 from qwen_vl_utils import process_vision_info
 import torch
 from PIL import Image
@@ -49,15 +49,15 @@ def array_to_image_path(image_array):
     return full_path
     
 models = {
-    "Qwen/Qwen2-VL-7B-Instruct": Qwen2VLForConditionalGeneration.from_pretrained("Qwen/Qwen2-VL-7B-Instruct", trust_remote_code=True, torch_dtype="auto").cuda().eval()
+    "Qwen2.5-VL-7B-Instruct": Qwen2_5_VLForConditionalGeneration.from_pretrained("Qwen2.5-VL-7B-Instruct", trust_remote_code=True, torch_dtype="auto").cuda().eval()
 
 }
 
 processors = {
-    "Qwen/Qwen2-VL-7B-Instruct": AutoProcessor.from_pretrained("Qwen/Qwen2-VL-7B-Instruct", trust_remote_code=True)
+    "Qwen2.5-VL-7B-Instruct": AutoProcessor.from_pretrained("Qwen2.5-VL-7B-Instruct", trust_remote_code=True)
 }
 
-DESCRIPTION = "This demo uses[Qwen2-VL-7B-Instruct](https://huggingface.co/Qwen/Qwen2-VL-7B-Instruct)"
+DESCRIPTION = "This demo uses[Qwen2.5-VL-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-VL-7B-Instruct)"
 
 kwargs = {}
 kwargs['torch_dtype'] = torch.bfloat16

requirements.txt

@@ -3,8 +3,6 @@ Pillow==10.3.0
 Requests==2.31.0
 torch
 torchvision
-git+https://github.com/huggingface/transformers.git
+git+https://github.com/huggingface/transformers
 accelerate
 qwen-vl-utils
-git+https://github.com/huggingface/transformers.git
-gliner