News
ResearchOpenAI pens AI safety agreement with four firmsReleasesOVHcloud unveils Olala mini, its first in-house LLMChinaA 9B model beats a 1T model on FinQA, trained with Crusoe Managed RLReleasesThe internal metaphysics of an LLMReleasesMapping Language Model Internal Representations to Physical Laws and Scientific ReasoningResearchCan China keep rogue AI in check? Meet the start-up testing the top modelsFrontierMicrosoft aims to challenge OpenAI, Anthropic with cheaper, specialized AI modelsOpen modelsMicrosoft rolls out new first-party models and updates to agentic AI continuous improvementFrontierOpenAI鈥檚 New Hacking Model Found Dozens Of Holes In Apple TechFrontierMicrosoft unveils AI models for coding, cybersecurity as it takes on Anthropic, OpenAIResearchOpenAI pens AI safety agreement with four firmsReleasesOVHcloud unveils Olala mini, its first in-house LLMChinaA 9B model beats a 1T model on FinQA, trained with Crusoe Managed RLReleasesThe internal metaphysics of an LLMReleasesMapping Language Model Internal Representations to Physical Laws and Scientific ReasoningResearchCan China keep rogue AI in check? Meet the start-up testing the top modelsFrontierMicrosoft aims to challenge OpenAI, Anthropic with cheaper, specialized AI modelsOpen modelsMicrosoft rolls out new first-party models and updates to agentic AI continuous improvementFrontierOpenAI鈥檚 New Hacking Model Found Dozens Of Holes In Apple TechFrontierMicrosoft unveils AI models for coding, cybersecurity as it takes on Anthropic, OpenAI
EveryLLM

Japanese

Swallow, PLaMo, Sarashina, LLM-jp. 15 models on EveryLLM so far.

Japanese models are built for Japanese, which mixes three scripts (kanji, hiragana and katakana), leaves no spaces between words and depends heavily on politeness levels and context. Japan has many builders: universities and national research institutes, telecoms and electronics groups, and start-ups such as Sakana AI and Preferred Networks, supported by a government programme that provides computing time.

Builders

Tokyo Institute of Technology and AIST built the Swallow models by continuing to train Llama models on Japanese text. Fugaku-LLM, a 13-billion-parameter model released in May 2024 by a team including Tokyo Tech, Tohoku University, Fujitsu and RIKEN, was trained on the Fugaku supercomputer with about 60% Japanese data. Preferred Networks builds PLaMo, NTT its tsuzumi models, SoftBank's SB Intuitions the Sarashina series, and the National Institute of Informatics leads LLM-jp.

Government and foreign labs

The Ministry of Economy, Trade and Industry launched the GENIAC programme in 2024 to give model builders computing time and support. Foreign labs court Japanese users too: OpenAI opened a Tokyo office in April 2024 with a GPT-4 model customised for Japanese, and Google released a Japanese version of Gemma 2 in October 2024. Sakana AI, founded in Tokyo in 2023, focuses on methods such as merging existing models into new ones.

Famous japanese models

Further reading

Swallow 馃嚡馃嚨 JapanSwallow Japanese-enhanced Llama models by Tokyo TechGeneral 路 Open weights 路 2024 Sarashina2 馃嚡馃嚨 JapanSarashina2 Open-weight Japanese and English language models by SB IntuitionsGeneral 路 Open weights RakutenAI-7B 馃嚡馃嚨 JapanRakutenAI-7B Japanese-focused open language model based on MistralGeneral 路 Open weights 路 2024 PLaMo-100B 馃嚡馃嚨 JapanPLaMo-100B A 100-billion parameter language model built for Japanese.General 路 Open weights 路 2024 EvoLLM-JP 馃嚡馃嚨 JapanEvoLLM-JP Experimental Japanese language model created by merging open-source models.General 路 Open weights 路 2024 PLaMo 2 馃嚡馃嚨 JapanPLaMo 2 Japanese-focused large language model series by Preferred NetworksGeneral 路 Open weights 路 2025 Tsuzumi 馃嚡馃嚨 JapanTsuzumi Lightweight Japanese LLM for business useGeneral 路 API only 路 2024 LLM-jp-3 馃嚡馃嚨 JapanLLM-jp-3 Open Japanese language model family by NIIGeneral 路 Open weights 路 2024 Llama-3-ELYZA-JP-8B 馃嚡馃嚨 JapanLlama-3-ELYZA-JP-8B Japanese-tuned Llama 3 model by ELYZAGeneral 路 Conditional licence Stockmark-100b 馃嚡馃嚨 JapanStockmark-100b A 100 billion parameter Japanese and English language model.General 路 Open weights Fugaku-LLM 馃嚡馃嚨 JapanFugaku-LLM Japanese large language model trained on Fugaku supercomputerGeneral 路 Open weights 路 2024 fnNemotron 3 Nano 馃嚭馃嚫 United StatesNemotron 3 Nano NVIDIA's 30B hybrid Mamba-Transformer MoE model for reasoning and agentsGeneral 路 Conditional licence 路 2025 CyberAgentLM3 馃嚡馃嚨 JapanCyberAgentLM3 A 22-billion parameter Japanese and English chat model by CyberAgent.General 路 Open weights

The weekly digest

Every Monday: the model of the day, new pages on EveryLLM, and the conferences and anniversaries of the week ahead.