Update app.py

app.py

@@ -1,15 +1,12 @@
 import gradio as gr
 from huggingface_hub import InferenceClient
 
-# Inference
-
-#model_text = "meta-llama/Llama-4-Scout-17B-16E-Instruct"
-model_text = "meta-llama/Llama-3.2-3B-Instruct"
-model_vision = "meta-llama/Llama-3.2-11B-Vision-Instruct"
-
 client = InferenceClient()
 
-def fn_text(
+# Llama 3 - Text
+model_llama_3_text = "meta-llama/Llama-3.2-3B-Instruct"
+
+def fn_llama_3_text(
     prompt,
     history,
     system_prompt,
@@ -18,14 +15,14 @@ def fn_text(
     top_p,
 ):
     
+    # With System Prompt
     messages = [{"role": "system", "content": [{"type": "text", "text": system_prompt}]}]
     history.append(messages[0])
-    
     messages.append({"role": "user", "content": [{"type": "text", "text": prompt}]})
     history.append(messages[1])
     
     stream = client.chat.completions.create(
-        model = model_text,
+        model = model_llama_3_text,
         messages = history,
         max_tokens = max_tokens,
         temperature = temperature,
@@ -39,7 +36,7 @@ def fn_text(
         yield "".join(chunks)
 
 app_text = gr.ChatInterface(
-    fn = fn_text,
+    fn = fn_llama_3_text,
     type = "messages",
     additional_inputs = [
         gr.Textbox(value="You are a helpful assistant.", label="System Prompt"),
@@ -47,11 +44,14 @@ app_text = gr.ChatInterface(
         gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
         gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-P"),
     ],
-    title = "Meta Llama",
-    description = model_text,
+    title = "Meta Llama 3",
+    description = model_llama_3_text,
 )
 
-def fn_vision(
+# Llama 3 - Vision
+model_llama_3_vision = "meta-llama/Llama-3.2-11B-Vision-Instruct"
+
+def fn_llama_3_vision(
     prompt,
     image_url,
     #system_prompt,
@@ -59,13 +59,14 @@ def fn_vision(
     temperature,
     top_p,
 ):
-    messages = [{"role": "user", "content": [{"type": "text", "text": prompt}]}]
     
+    # Without System Prompt
+    messages = [{"role": "user", "content": [{"type": "text", "text": prompt}]}]
     if image_url:
         messages[0]["content"].append({"type": "image_url", "image_url": {"url": image_url}})
     
     stream = client.chat.completions.create(
-        model = model_vision,
+        model = model_llama_3_vision,
         messages = messages,
         max_tokens = max_tokens,
         temperature = temperature,
@@ -78,8 +79,8 @@ def fn_vision(
         chunks.append(chunk.choices[0].delta.content or "")
         yield "".join(chunks)
 
-app_vision = gr.Interface(
-    fn = fn_vision,
+app_llama_3_vision = gr.Interface(
+    fn = fn_llama_3_vision,
     inputs = [
         gr.Textbox(label="Prompt"),
         gr.Textbox(label="Image URL")
@@ -93,14 +94,11 @@ app_vision = gr.Interface(
         gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
         gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-P"),
     ],
-    title = "Meta Llama",
-    description = model_vision,
+    title = "Meta Llama 3",
+    description = model_llama_3_vision,
 )
 
 app = gr.TabbedInterface(
-    [app_text, app_vision],
-    ["Text", "Vision"]
-).launch()
-
-#if __name__ == "__main__":
-#    app.launch()
+    [app_llama_3_text, app_llama_3_vision],
+    ["Llama 3 - Text", "Llama 3 - Vision"]
+).launch()