Liquid AI releases early checkpoint LFM2-24B-A2B, 24B total parameters with only 2.3B active
Original title:LFM2-24B-A2B: Scaling Up the LFM2 Architecture
AI summary
Liquid AI releases an early checkpoint of LFM2-24B-A2B, the largest model in the LFM2 series, with an MoE architecture, 24B total parameters and about 2.3B active per token, deployable in 32GB of memory on cloud and edge devices.
Key Takeaways
- Liquid AI releases an early checkpoint of LFM2-24B-A2B,the largest model in the LFM2 series
- with an MoE architecture,24B total parameters and about 2.
- 3B active per token,deployable in 32GB of memory on cloud and edge devices.
Quick Facts
- Published
- Updated
- Source
- Liquid AI 模型与工程博客(网页)
- Category
- Model Releases
- AI score
- 64
官方给出架构配方和与 Qwen3、gpt-oss 的吞吐对比,读者可以据此评估这款 24B MoE 在笔记本和边缘部署上的实际可行性。
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: Liquid AI releases early checkpoint LFM2-24B-A2B, 24B total parameters with only 2.3B active. February 24, 2026. https://new.nat6.net/en/news/cmtym3yoy000dronzcgp7p57y/