Update app.py

app.py

@@ -12,24 +12,24 @@ import torch
 import gradio as gr
 from gradio.themes.utils import colors, fonts, sizes
 
-from faster_whisper import WhisperModel
+# from faster_whisper import WhisperModel
 from moviepy.editor import VideoFileClip
 from transformers import AutoTokenizer, AutoModel
 
 import subprocess
-subprocess.run('pip install flash-attn --no-build-isolation', env={'FLASH_ATTENTION_SKIP_CUDA_BUILD': "TRUE"}, shell=True)
+# subprocess.run('pip install flash-attn --no-build-isolation', env={'FLASH_ATTENTION_SKIP_CUDA_BUILD': "TRUE"}, shell=True)
 
 # ========================================
 #             Model Initialization
 # ========================================
 
 if gr.NO_RELOAD:
-    if torch.cuda.is_available():
-        speech_model = WhisperModel("large-v3", device="cuda", compute_type="float16")
-    else:
-        speech_model = WhisperModel("large-v3", device="cpu")
+    # if torch.cuda.is_available():
+        # speech_model = WhisperModel("large-v3", device="cuda", compute_type="float16")
+    # else:
+        # speech_model = WhisperModel("large-v3", device="cpu")
 
-    model_path = 'OpenGVLab/VideoChat-Flash-Qwen2-7B_res448'
+    model_path = 'OpenGVLab/InternVideo2_5_Chat_8B'
 
     tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True)
     model = AutoModel.from_pretrained(model_path, trust_remote_code=True).half().cuda()