CyberAgent
1 model across 1 family · Latest: Open-CALM 7B (2023-05)
Japanese language models for digital advertising.
CyberAgent is a Japanese AI company founded in 1998. Japanese language models for digital advertising. CyberAgent ships 1 model family totaling 1 model, with the most recent release Open-CALM 7B in 2023-05. Notable families include Open-CALM. Use it as a stable reference for lab background, release coverage, and follow-up model pages as they are added. Researchers and evaluators can scan counts, links, release history, and source references without leaving the directory.
About
CyberAgent is a major Japanese internet company that developed the OpenCALM series of decoder-only language models pre-trained on Japanese datasets, ranging from 160M to 6.8B parameters. The successor CALM2 was trained on 1.3 trillion tokens of Japanese and English data, and CALM3 on 2.0 trillion tokens, with a 22B-parameter chat variant specialized for dialogue. CyberAgent's AI Lab ranks 4th domestically and 49th globally among AI research organizations.
Featured models
| Model | Released | Context | Input price ($/1M) | Output price ($/1M) | License |
|---|---|---|---|---|---|
| Open-CALM 7B | 2023-05-01 | - | - | - | CC-BY-SA-4.0 |
Model families
Recent releases
- Open-CALM 7B- 2023-05-01
FAQ
Who founded CyberAgent and when?
CyberAgent was founded in 1998 and is associated with Tokyo, Japan.
What models has CyberAgent released?
CyberAgent ships 1 model across 1 family: Open-CALM.
Is CyberAgent's technology open source?
All tracked models are released under CC-BY-SA-4.0.
Where is CyberAgent headquartered?
CyberAgent is headquartered in Tokyo, Japan.
What is CyberAgent known for?
Japanese language models for digital advertising. Its most prominent tracked family is Open-CALM.
How can I access CyberAgent's models?
CyberAgent's provider availability is tracked on model pages as API and hosting data is verified.
Explore related pages
Last reviewed: 2026-04-18. Data sourced from public lab announcements and provider documentation.