Update app.py

app.py

@@ -1,27 +1,37 @@
 import gradio as gr
-from transformers import AutoProcessor, AutoModelForVisionLanguage
 import torch
+from transformers import AutoProcessor, LlavaForConditionalGeneration
 
 MODEL_NAME = "Declan1/llava-v1.6-mistral-7b-sydneyfish-a100"
 
+# Load processor (handles image + text formatting)
 processor = AutoProcessor.from_pretrained(MODEL_NAME)
-model = AutoModelForVisionLanguage.from_pretrained(
+
+# Load model (Vision + Language fused)
+model = LlavaForConditionalGeneration.from_pretrained(
     MODEL_NAME,
     torch_dtype=torch.float16,
     device_map="auto"
 )
 
-def chat(image, text):
-    inputs = processor(text=text, images=image, return_tensors="pt").to(model.device)
+def chat(image, prompt):
+    # Process inputs
+    inputs = processor(prompt, images=image, return_tensors="pt").to(model.device)
+    
+    # Generate output
     output = model.generate(**inputs, max_new_tokens=300)
+    
+    # Decode to readable text
     response = processor.decode(output[0], skip_special_tokens=True)
     return response
 
-iface = gr.Interface(
+# Simple UI
+interface = gr.Interface(
     fn=chat,
-    inputs=[gr.Image(type="pil"), gr.Textbox(label="Prompt")],
+    inputs=[gr.Image(type="pil", label="Image"), gr.Textbox(label="Prompt")],
     outputs=gr.Textbox(label="Response"),
-    title="SydneyFish LLaVA 1.6 - Mistral 7B"
+    title="SydneyFish LLaVA (Mistral 7B)",
+    description="Upload an image and ask it something."
 )
 
-iface.launch()
+interface.launch()