{"paper":{"title":"K-EXAONE 2.0 Technical Report","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ahra Jo, Byungoh Ko, Chaeeun Lee, ChaeYoon Lee, Chang En Yea, Changhun Lee, Chansik Yoon, Chanwoo Moon, Daeseong Kim, Dasol Hong, Dohaeng Lee, Donghyeon Shin, Donghyun Son, Doyoung Kim, Edward Hwayoung Lee, Euisoon Kim, Eunbi Choi, Eunyong Son, Gerrard Jeongwon Jo, Haeju Lee, Hayeon Kim, Heegyu Kim, Heuiyeen Yeen, Honglak Lee, Hwansoo Lee, Hyeongjun Jo, Hyerin Seo, Hyojin Jeon, Hyosang Kim, Hyunjik Jo, Jimin Park, Jinsik Lee, Jiwon Ham, Jiyeon Jung, Joonkee Kim, Jueun Mun, Junwon Hwang, Kaehyun Um, Kibong Choi, Kijeong Jeon, Kijung Seo, Kwangrok Ryoo, Kyungjae Yoo, Kyungmin Lee, Minhyeok Jung, Minju Seo, Minkyung Lee, Minwoo Lee, Myoungshin Kim, Naeun Kang, Sangeun Lee, Sangha Park, Sehyun Chun, Sejong Yang, Seojeong Park, Seokhee Hong, Seonghwan Kim, Sihoon Yang, Sihyuk Yi, Solji Lim, Soyeon Kim, Stanley Jungkyu Choi, Sujeong Im, Sungjune Park, Sunkyoung Kim, Unsol Kim, Wonkee Lee, Woohyung Lim, Yemuk Choi, Yeonsik Jo, Yireun Kim, Yongchan Chun, Yongil Kim, Yongmin Park, Yongwoo Song, Youchul Kim, Yujin Jo","submitted_at":"2026-08-05T06:41:48Z","abstract_excerpt":"This technical report presents K-EXAONE 2.0, an open-weight multilingual foundation model developed by LG AI Research as a step in our effort toward global frontier-scale foundation models. Rather than training from scratch, we upcycle K-EXAONE and expand its architecture, yielding a Mixture-of-Experts (MoE) model with 750B total parameters and approximately 37B activated per token---more than three times the capacity of its predecessor. K-EXAONE 2.0 supports context lengths of up to 256K tokens and expands multilingual coverage from six to ten languages. Its training pipeline combines continu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.04505","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2608.04505/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}