Kimi releases and open-sources K2.5 model with new visual understanding, coding, and Agent cluster capabilities
AI summary
Kimi released and open-sourced K2.5, its most intelligent and versatile model to date. Its native multimodal architecture supports visual and text input and thinking and non-thinking modes, achieving open-source state-of-the-art performance on Agent, coding, image, video, and general intelligence tasks.
Key Takeaways
- Kimi released and open-sourced K2.
- 5,its most intelligent and versatile model to date.
- Its native multimodal architecture supports visual and text input and thinking and non-thinking…
Quick Facts
- Published
- Updated
- Source
- 公众号:月之暗面(Kimi)
- Category
- Model Releases
- AI score
- 61
Kimi K2.5把视觉理解和Agent集群能力打包开源,国内模型里比较早做这个组合的,虽然参数规模争议不小,但对想搭多模态Agent的团队是个可用的起点。
Related entities
Source & copyright
This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.
How to Cite
When citing this article, keep the title, date, and permalink, and credit Orange News as the source.
Orange News: Kimi releases and open-sources K2.5 model with new visual understanding, coding, and Agent cluster capabilities. January 27, 2026. https://new.nat6.net/en/news/cmq2otyj101rhsl6nuswe40m9/