Update app.py

app.py

@@ -25,7 +25,13 @@ class StopOnTokens(StoppingCriteria):
 @spaces.GPU
 def predict(message, history):
     stop = StopOnTokens()
-    prompt = tokenizer.apply_chat_template(history + message, tokenize=False, add_generation_prompt=True)
+    conversation = []
+    
+    for user, assistant in history:
+        conversation.extend([{"role": "user", "content": user}, {"role": "assistant", "content": assistant}])
+        
+    conversation.append({"role": "user", "content": message})
+    prompt = tokenizer.apply_chat_template(conversation, tokenize=False, add_generation_prompt=True)
     model_inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
     streamer = TextIteratorStreamer(tokenizer, timeout=10., skip_prompt=True, skip_special_tokens=True)
     generate_kwargs = dict(
@@ -52,7 +58,7 @@ def predict(message, history):
 
 # Setting up the Gradio chat interface.
 gr.ChatInterface(predict,
-                 title="Fialka 13B v4",
-                 description="Внимание! Все ответы сгенерированы и могут содержать неточную информацию.",
-                 examples=['Как приготовить рыбу?', 'Кто президент США?']
+                 title="SOLAR 10.7B Instruct v1.0",
+                 description="Warning. All answers are generated and may contain inaccurate information.",
+                 examples=['How do you cook fish?', 'Who is the president of the United States?']
                  ).launch()  # Launching the web interface.