Update app.py

app.py

@@ -9,7 +9,7 @@ quant = os.getenv('QUANT')
 chat_template = os.getenv('CHAT_TEMPLATE')
 
 # Interface variables
-model_name = model_id.split('/')[1].split('-GGUF')[0]
+model_name = model_id.split('/')[-1]
 title = f"🇩🇪 {model_name}"
 description = f"Chat with <a href=\"https://huggingface.co/{model_id}\">{model_name}</a> in GGUF format ({quant})!"
 
@@ -21,7 +21,8 @@ llm = Llama(model_path="model.gguf",
 
 # Function for streaming chat completions
 def chat_stream_completion(message, history, system_prompt):
-    messages_prompts = [{"role": "system", "content": system_prompt}]
+    #messages_prompts = [{"role": "system", "content": system_prompt}]
+    messages_prompts = []
     for human, assistant in history:
         messages_prompts.append({"role": "user", "content": human})
         messages_prompts.append({"role": "assistant", "content": assistant})
@@ -43,8 +44,8 @@ gr.ChatInterface(
     fn=chat_stream_completion,
     title=title,
     description=description,
-    additional_inputs=[gr.Textbox("Du bist ein hilfreicher Assistent.")],
-    additional_inputs_accordion="📝 System prompt",
+    #additional_inputs=[gr.Textbox("Du bist ein hilfreicher Assistent.")],
+    #additional_inputs_accordion="📝 System prompt",
     examples=[
         ["Was ist ein Large Language Model?"],
         ["Was ist 9+2-1?"],