{"@context":"https://neupai.io/schema/v0.2","@type":"StructuredNewsArticle","identity":{"article_id":"tech42_20260604_ai-pc-competition-system-optimization-nota-gpu-npu","canonical_url":"https://www.tech42.co.kr/ai-pc-%ea%b2%bd%ec%9f%81-%ec%b9%a9-%ec%84%b1%eb%8a%a5-%eb%84%98%ec%96%b4-%ec%8b%9c%ec%8a%a4%ed%85%9c-%ec%b5%9c%ec%a0%81%ed%99%94%eb%a1%9c%eb%85%b8%ed%83%80-gpu%c2%b7npu/?utm_source=rss&utm_medium=rss&utm_campaign=ai-pc-%25ea%25b2%25bd%25ec%259f%2581-%25ec%25b9%25a9-%25ec%2584%25b1%25eb%258a%25a5-%25eb%2584%2598%25ec%2596%25b4-%25ec%258b%259c%25ec%258a%25a4%25ed%2585%259c-%25ec%25b5%259c%25ec%25a0%2581%25ed%2599%2594%25eb%25a1%259c%25eb%2585%25b8%25ed%2583%2580-gpu%25c2%25b7npu","ai_url":null,"publisher":{"name":"테크42","domain":"www.tech42.co.kr","type":"online"},"author":"김한수 기자","published_at":"2026-06-04T00:33:53.000Z","updated_at":null,"language":"en","article_type":"straight_news","originality":"self_produced"},"content":{"headline":"AI PC Competition Goes Beyond Chip Performance to 'System Optimization'... Nota Implements Separate GPU-NPU Inference","summary":"Nota announced that it has implemented technology to run LLMs by separating GPU and NPU on Intel Lunar Lake-based AI PCs. The company achieved a 32% reduction in energy consumption compared to single GPU and an 89% reduction in first response latency compared to single NPU.","topics":["ai","semiconductor","technology"],"geography":["KR"],"entities":[{"name":"Nota","canonical_id":"corp:kr:nota","type":"company","role_in_article":"primary_subject","metadata":{"ticker":null,"parent":null}},{"name":"Intel","canonical_id":"corp:us:intel","type":"company","role_in_article":"mentioned","metadata":{"ticker":null,"parent":null}},{"name":"Chae Myung-soo","canonical_id":"person:kr:chae-myeong-soo","type":"person","role_in_article":"quoted","metadata":{"ticker":null,"parent":null}}],"claims":[{"id":"c1","statement":"Energy consumption per token was reduced by approximately 32% compared to single GPU execution method","as_of":"2026-06","as_of_explicit":false,"as_of_raw":"June 2026","source_type":"company_disclosure","comparison":"single_gpu_execution","type":"fact","figures":null,"expiry_hint":null,"insight":null},{"id":"c2","statement":"Generation throughput increased by approximately 12% compared to single GPU execution method","as_of":"2026-06","as_of_explicit":false,"as_of_raw":"June 2026","source_type":"company_disclosure","comparison":"single_gpu_execution","type":"fact","figures":null,"expiry_hint":null,"insight":null},{"id":"c3","statement":"First response latency was reduced by approximately 89% compared to single NPU method","as_of":"2026-06","as_of_explicit":false,"as_of_raw":"June 2026","source_type":"company_disclosure","comparison":"single_npu_execution","type":"fact","figures":null,"expiry_hint":null,"insight":null},{"id":"c4","statement":"Evaluation was conducted under conditions of Intel Lunar Lake-based AI PC, MoE LLM, and context window size of 512","as_of":"2026-06","as_of_explicit":false,"as_of_raw":"June 2026","source_type":"company_disclosure","comparison":null,"type":"fact","figures":null,"expiry_hint":null,"insight":null}],"ai_emotional_context":{"valence":0,"arousal":0,"primary_emotions":[],"secondary_emotions":[],"emotional_triggers":[]},"image":{"url":"https://tech42.shoplic.cloud/wp-content/uploads/2026/06/665394647-1024x480.png","alt":"기사 대표 이미지","caption":"동일한 LLM을 AI PC에서 실행한 비교 화면. 노타의 분리형 추론 방식은 입력 처리와 답변 생성을 GPU·NPU에 나눠 배치해 단일 연산 장치 기반 실행보다 빠르게 응답을 완료했다. (이미지=노타)","source":"first_img","alt_status":"auto"}},"provenance":{"source_chain":["primary_reporting"],"original_source_url":null,"related_articles":[]},"temporal":{"freshness":"recent","next_update_expected":null},"access":{"license":"neupai_standard","attribution_required":true,"structured_data":"free","full_text_available":false,"full_text_access":null}}