Update app.py

app.py

@@ -1,5 +1,6 @@
 from huggingface_hub import InferenceClient
 import gradio as gr
+import random
 
 API_URL = "https://api-inference.huggingface.co/models/"
 
@@ -7,7 +8,6 @@ client = InferenceClient(
     "mistralai/Mistral-7B-Instruct-v0.1"
 )
 
-
 def format_prompt(message, history):
   prompt = "<s>"
   for user_prompt, bot_response in history:
@@ -16,9 +16,7 @@ def format_prompt(message, history):
   prompt += f"[INST] {message} [/INST]"
   return prompt
 
-def generate(
-    prompt, history, temperature=0.9, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0,
-):
+def generate(prompt, history, temperature=0.9, max_new_tokens=512, top_p=0.95, repetition_penalty=1.0):
     temperature = float(temperature)
     if temperature < 1e-2:
         temperature = 1e-2
@@ -30,7 +28,7 @@ def generate(
         top_p=top_p,
         repetition_penalty=repetition_penalty,
         do_sample=True,
-        seed=42,
+        seed=random.randint(0, 10**7),
     )
 
     formatted_prompt = format_prompt(prompt, history)
@@ -56,9 +54,9 @@ additional_inputs=[
     ),
     gr.Slider(
         label="Max new tokens",
-        value=256,
+        value=512,
         minimum=64,
-        maximum=8000,
+        maximum=1024,
         step=64,
         interactive=True,
         info="The maximum numbers of new tokens",
@@ -83,7 +81,24 @@ additional_inputs=[
     )
 ]
 
-with gr.Blocks() as demo:
+customCSS = """
+.contain {
+  display: flex;
+  flex-direction: column;
+}
+.gradio-container {
+  height: 100vh !important;
+}
+#component-0 {
+  height: 100%;
+}
+#chatbot {
+  flex-grow: 1;
+  overflow: auto;
+}
+"""
+
+with gr.Blocks(css=customCSS) as demo:
     gr.ChatInterface(
         generate,
         additional_inputs=additional_inputs,