Update app.py

app.py

@@ -1,19 +1,18 @@
 import gradio as gr
+import torch
 from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
 import threading
-import torch
 
-# Detectar dispositivo automaticamente (GPU ou CPU)
+# Detectar dispositivo automaticamente
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 
-# Carregar modelo e tokenizer
+# Inicializar o modelo e o tokenizer
 model_name = "lambdaindie/lambda-1v-1B"
 model = AutoModelForCausalLM.from_pretrained(model_name).to(device)
 tokenizer = AutoTokenizer.from_pretrained(model_name)
 
 stop_flag = {"stop": False}
 
-# Função de resposta
 def respond(prompt, history):
     stop_flag["stop"] = False
 
@@ -22,7 +21,6 @@ def respond(prompt, history):
 
     streamer = TextIteratorStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)
 
-    # Iniciar thread de geração
     generation_thread = threading.Thread(
         target=model.generate,
         kwargs={
@@ -43,29 +41,26 @@ def respond(prompt, history):
         if stop_flag["stop"]:
             return "", history
         reasoning += new_text
-        yield "", (history or []) + [(prompt, f"<div class='final-answer'>{reasoning}</div>")]
+        yield "", history[:-1] + [(prompt, f"<div class='final-answer'>{reasoning}</div>")]
 
-# Função para parar a geração
 def stop_generation():
     stop_flag["stop"] = True
 
-# Interface Gradio
 with gr.Blocks(css="""
-    #chatbot, .gr-markdown, .gr-button, .gr-textbox {
-        font-family: 'JetBrains Mono', monospace !important;
-        font-size: 11px !important;
-    }
-    .final-answer {
-        background-color: #1e1e1e;
-        color: #ffffff;
-        padding: 10px;
-        border-left: 4px solid #4caf50;
-        font-family: 'JetBrains Mono', monospace !important;
-        white-space: pre-wrap;
-        font-size: 11px !important;
+    #chatbot, .gr-markdown, .gr-button, .gr-textbox {  
+        font-family: 'JetBrains Mono', monospace !important;  
+        font-size: 11px !important;  
+    }  
+    .final-answer {  
+        background-color: #1e1e1e;  
+        color: #ffffff;  
+        padding: 10px;  
+        border-left: 4px solid #4caf50;  
+        font-family: 'JetBrains Mono', monospace !important;  
+        white-space: pre-wrap;  
+        font-size: 11px !important;  
     }
 """) as demo:
-    gr.Markdown('<link href="https://fonts.googleapis.com/css2?family=JetBrains+Mono&display=swap" rel="stylesheet">')
     gr.Markdown("## λambdAI — Reasoning Chat")
 
     chatbot = gr.Chatbot(elem_id="chatbot")