Update app.py

app.py

@@ -27,8 +27,8 @@ MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "4096"))
 
 device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
 
-# Load Vision-Matters-7B
-MODEL_ID_M = "Yuting6/Vision-Matters-7B"
+# Load Camel-Doc-OCR-080125
+MODEL_ID_M = "prithivMLmods/Camel-Doc-OCR-080125"
 processor_m = AutoProcessor.from_pretrained(MODEL_ID_M, trust_remote_code=True)
 model_m = Qwen2_5_VLForConditionalGeneration.from_pretrained(
     MODEL_ID_M, trust_remote_code=True,
@@ -101,7 +101,7 @@ def generate_image(model_name: str,
     """
     Generates responses using the selected model for image input.
     """
-    if model_name == "Vision-Matters-7B":
+    if model_name == "Camel-Doc-OCR-080125(v2)":
         processor = processor_m
         model = model_m
     elif model_name == "ViGaL-7B":
@@ -168,7 +168,7 @@ def generate_video(model_name: str,
     """
     Generates responses using the selected model for video input.
     """
-    if model_name == "Vision-Matters-7B":
+    if model_name == "Camel-Doc-OCR-080125(v2)":
         processor = processor_m
         model = model_m
     elif model_name == "ViGaL-7B":
@@ -331,14 +331,14 @@ with gr.Blocks(css=css, theme="bethecloud/storj_theme") as demo:
                 #download_btn = gr.Button("Download Result.md")
 
             model_choice = gr.Radio(choices=[
-                 "Vision-Matters-7B", "WR30a-Deep-7B-0711",
+                 "Camel-Doc-OCR-080125(v2)", "WR30a-Deep-7B-0711",
                  "ViGaL-7B", "MonkeyOCR-pro-1.2B", "Visionary-R1-3B"
             ],
                                     label="Select Model",
                                     value="Vision-Matters-7B")
 
             gr.Markdown("**Model Info 💻** | [Report Bug](https://huggingface.co/spaces/prithivMLmods/Multimodal-VLMs-5x/discussions)")         
-            gr.Markdown("> [WR30a-Deep-7B-0711](https://huggingface.co/prithivMLmods/WR30a-Deep-7B-0711): wr30a-deep-7b-0711 model is a fine-tuned version of qwen2.5-vl-7b-instruct, optimized for image captioning, visual analysis, and image reasoning. Built on top of the qwen2.5-vl architecture, this experimental model enhances visual comprehension capabilities with focused training on 1,500k image pairs for superior image understanding.")
+            gr.Markdown("> [Camel-Doc-OCR-080125(v2)](https://huggingface.co/prithivMLmods/WR30a-Deep-7B-0711): the camel-doc-ocr-080125 model is a fine-tuned version of qwen2.5-vl-7b-instruct, optimized for document retrieval, content extraction, and analysis recognition. built on top of the qwen2.5-vl architecture, this model enhances document comprehension capabilities.")
             gr.Markdown("> [MonkeyOCR-pro-1.2B](https://huggingface.co/echo840/MonkeyOCR-pro-1.2B): MonkeyOCR adopts a structure-recognition-relation (SRR) triplet paradigm, which simplifies the multi-tool pipeline of modular approaches while avoiding the inefficiency of using large multimodal models for full-page document processing.")
             gr.Markdown("> [Vision Matters 7B](https://huggingface.co/Yuting6/Vision-Matters-7B): vision-matters is a simple visual perturbation framework that can be easily integrated into existing post-training pipelines including sft, dpo, and grpo. our findings highlight the critical role of visual perturbation: better reasoning begins with better seeing.")
             gr.Markdown("> [ViGaL 7B](https://huggingface.co/yunfeixie/ViGaL-7B): vigal-7b shows that training a 7b mllm on simple games like snake using reinforcement learning boosts performance on benchmarks like mathvista and mmmu without needing worked solutions or diagrams indicating transferable reasoning skills.")