Krafton Unveils Preview of In-House Developed LLM 'Raon-1 Mini'

조선일보

KRAFTON | 크래프톤 공식 웹사이트

View article

Reposted by

KRAFTON | 크래프톤 공식 웹사이트

Lean Right

Factuality

Ownership

Krafton announced on the 21st that it has released a preview version of its self-developed high-efficiency massive language model (LLM), ‘Raon-1 mini.’ Raon-1 mini is an LLM with 30 billion (30B) parameters, for which Krafton independently carried out the entire process from model design to training. It supports Korean and English and focuses on simultaneously achieving high performance and low operating costs.

연합뉴스-Yonhap News Agency

Lean Right

Factuality

Ownership

(Seoul = Yonhap News) Reporter Kim Ju-hwan = Krafton [259960] [is] of its self-developed 30B-class Large Language Model (LLM) 'Raon-1 mini'...

asiae.co.kr

Factuality

Ownership

Krafton announced on the 21st that it has released a preview version of its self-developed 30B-class massive language model (LLM), 'Raon-1 Mini'.

ddaily.co.kr

Factuality

Ownership

[Digital Daily Reporter Lee Hak-beom] Krafton has unveiled a Massive Language Model (LLM) that it developed in-house, from design to training. By utilizing only a fraction of the total parameters, it reduces the cost of AI inference and has released the model to allow external developers to use it commercially. Krafton announced on the 21st that it has released a preview version of its in-house developed LLM, 'Raon-1 Mini,' featuring 30 billion…