Handle correctly multiple images

app.py

@@ -48,7 +48,8 @@ def run_inference(model_key, input_type, text, image, video, fps, system_prompt,
     # Multimodal inference (image or video)
     content = []
     if input_type == "image" and image:
-        content.append({"type": "image", "image": image})
+        for img_path in image:
+            content.append({"type": "image", "image": img_path})
     elif input_type == "video" and video:
         # Ensure file URI for local files
         video_src = video if str(video).startswith("file://") else f"file://{video}"