Update app.py

app.py

@@ -56,23 +56,19 @@ def generate(image, video, audio, instruction=DEFAULT_INPUT, sampling=False, tem
     if image is not None:
         img = Image.fromarray(image.astype("uint8"), "RGB")
         inputs = [{"role": "user", "content": [img, instruction]}]
-        media = img
     elif video is not None:
         vid = encode_video(video)
-        inputs = [{"role": "user", "content": [vid, instruction]}]
-        media = vid
+        inputs = [{"role": "user", "content": vid + [instruction]}]
     elif audio is not None:
         if isinstance(audio, str):
             aud, _ = librosa.load(audio, sr=16000, mono=True)
         else:
             aud = audio
         inputs = [{"role": "user", "content": [aud, instruction]}]
-        media = aud
     else:
         return "No input provided."
 
     params = {
-        "image": media,
         "msgs": inputs,
         "tokenizer": tokenizer,
         #"processor": processor,