Update app.py

app.py

@@ -7,16 +7,12 @@ from huggingface_hub import InferenceClient
 # Initialize the Hugging Face Inference Client
 client = InferenceClient("microsoft/llava-med-7b-delta")
 
-# Custom Field for Base64 Encoded Image
-class Base64ImageField(gr.Field):
-    def __init__(self, *args, **kwargs):
-        super().__init__(*args, **kwargs)
-
-    def preprocess(self, image):
-        buffered = io.BytesIO()
-        image.save(buffered, format="PNG")
-        img_str = base64.b64encode(buffered.getvalue()).decode('utf-8')
-        return img_str
+# Function to encode image as base64 (optional if Gradio handles image conversion)
+def image_to_base64(image):
+    buffered = io.BytesIO()
+    image.save(buffered, format="PNG")
+    img_str = base64.b64encode(buffered.getvalue()).decode('utf-8')
+    return img_str
 
 # Function to interact with LLAVA model
 def respond(
@@ -39,14 +35,8 @@ def respond(
     messages.append({"role": "user", "content": message})
 
     if image:
-        # Convert image(s) to base64 using the custom field
-        if isinstance(image, Image.Image):
-            image_b64 = Base64ImageField().preprocess(image)
-            messages.append({"role": "user", "content": "Image uploaded", "image": image_b64})
-        else:
-            for img in image:
-                image_b64 = Base64ImageField().preprocess(img)
-                messages.append({"role": "user", "content": "Image uploaded", "image": image_b64})
+        # Gradio handles image processing internally, so no need for manual base64 encoding
+        messages.append({"role": "user", "content": "Image uploaded"})
 
     # Call Hugging Face model for response
     try: