from nzfc_gram_runtime import NZFCGramLongMemoryChat from nzfc_gram_runtime.nonquant import attach_nonquant_gemma, patch_generation_use_cache_false bot = NZFCGramLongMemoryChat(repo_dir='.', load_model=False, require_model=False) attach_nonquant_gemma(bot, gpu_max_memory_gib=11, device_map='balanced_low_0') patch_generation_use_cache_false(bot) out = bot.chat( 'NZFC-GRAM\uc758 \uc7a5\uae30\uae30\uc5b5 \uacbd\uacc4\uac00 \ub0b4\ubd80 context\uc778\uc9c0 \uc678\ubd80 memory retrieval\uc778\uc9c0 \uc124\uba85\ud574\uc918.', user_id='quick_user', project_id='quick_project', session_id='quick_session', save_turn=False, max_new_tokens=80, ) print(out['answer'])