Update app/llm.py

app/llm.py

@@ -34,7 +34,7 @@ llm_chat = llama_cpp.Llama.from_pretrained(
     filename="healthcare-ai-q8_0-unsloth.Q8_0.gguf",
     #tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained("moriire/healthcare-ai-q8_0"),
     verbose=False,
-     n_ctx=1024,
+     n_ctx=256,
      n_gpu_layers=0,
     #chat_format="llama-2"
 )