Update app.py

app.py

@@ -34,11 +34,6 @@ hf_hub_download(
 # Set the title and description
 title = "Dolphin-3 Llama.cpp 🐬"
 description = """Dolphin 3.0 is a powerful, general-purpose local AI model designed for coding, math, and various other tasks, aiming similar to the models like ChatGPT and Claude."""
-# examples = [
-#     ["Provide a historical fact about the capital of France."],
-#     ["Suggest a three-day itinerary for exploring Tokyo and nearby areas?"],
-#     ["Can you write a short, imaginative story about exploring a forest?"],
-# ]
 
 
 def respond(
@@ -134,51 +129,68 @@ demo = gr.ChatInterface(
     ),
     additional_inputs=[
         gr.Dropdown(
-            [
+            choices=[
                 "Dolphin3.0-Llama3.2-1B-Q6_K.gguf",
                 "Dolphin3.0-Qwen2.5-1.5B-Q6_K.gguf",
                 "Dolphin3.0-Qwen2.5-0.5B-Q6_K.gguf",
             ],
             value="Dolphin3.0-Llama3.2-1B-Q6_K.gguf",
             label="Model",
+            info="Select the AI model to use for chat"
         ),
         gr.Textbox(
-            value="You are Dolphin an AI assistant that helps humanity.",
-            label="System message",
+            value="You are Dolphin, a helpful AI assistant focused on accurate and ethical responses.",
+            label="System Prompt",
+            info="Define the AI assistant's personality and behavior",
+            lines=2
+        ),
+        gr.Slider(
+            minimum=512,
+            maximum=8192,
+            value=4096,
+            step=512,
+            label="Max Tokens",
+            info="Maximum length of response (higher = longer replies)"
+        ),
+        gr.Slider(
+            minimum=0.1,
+            maximum=2.0,
+            value=0.7,
+            step=0.1,
+            label="Temperature",
+            info="Creativity level (higher = more creative, lower = more focused)"
         ),
-        gr.Slider(minimum=1, maximum=8192, value=8192, step=1, label="Max tokens"),
-        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",
+            info="Nucleus sampling threshold"
         ),
         gr.Slider(
-            minimum=0,
+            minimum=1,
             maximum=100,
             value=40,
             step=1,
             label="Top-k",
+            info="Limit vocabulary choices to top K tokens"
         ),
         gr.Slider(
-            minimum=0.0,
+            minimum=1.0,
             maximum=2.0,
             value=1.1,
             step=0.1,
-            label="Repetition penalty",
-        ),
-    ],
-    theme="Glass",
+            label="Repetition Penalty",
+            info="Penalize repeated words (higher = less repetition)"
+        )
+    ]
+    theme="Base",
     submit_btn="Send",
     stop_btn="Stop",
     title=title,
     description=description,
     chatbot=gr.Chatbot(scale=1, show_copy_button=True),
-    # examples=examples,
-    cache_examples=True,
-    cache_mode="lazy",
     flagging_mode="never",
 )