Update app.py

app.py

@@ -7,16 +7,17 @@ multiprocessing.set_start_method("spawn", force=True)
 
 # --- Trending models for image text-to-text tasks ---
 TRENDING_MODELS = [
-    "Salesforce/blip2-opt-2.7b",
-    "Salesforce/blip2-flan-t5-xl",
-    "Salesforce/blip-image-captioning-base",
-    "Salesforce/blip-image-captioning-large",
-    "nlpconnect/vit-gpt2-image-captioning",
-    "OFA-Sys/OFA-base",
-    "OFA-Sys/OFA-large",
-    "dandelin/vilt-b32-finetuned-vqa",
-    "dandelin/vilt-b32-mlm",
-    "uclanlp/visualbert-vqa-coco-pre"
+    "Salesforce/blip2-opt-2.7b",             # Uses Blip2Config
+    "Salesforce/blip2-flan-t5-xl",             # Uses Blip2Config
+    "Salesforce/instructblip-vicuna-7b",       # Uses InstructBlipConfig
+    "LLaVA/LLaVA-7B",                        # Uses LlavaConfig
+    "LLaVA/LLaVA-13B",                       # Uses LlavaConfig
+    "LlavaNext/LlavaNext-7B",                  # Uses LlavaNextConfig
+    "Qwen2-VL-7B",                           # Uses Qwen2VLConfig
+    "google/pix2struct-base",                # Uses Pix2StructConfig
+    "nlpconnect/vit-gpt2-image-captioning",  # Uses VisionEncoderDecoderConfig
+    "LlavaOneVision/LlavaOneVision-7B",        # Uses LlavaOnevisionConfig
+    "MosaicML/mllama-7b"                     # Uses MllamaConfig
 ]
 
 # --- Helper: if the user selects "Custom", then they can enter any model identifier ---