**LFM2.5-Encoders: Fast at Long Context, Even on CPU**

> Open-weight bidirectional encoders for classification, routing & NLU. LFM2.5-Encoder-230M & 350M deliver 8K context and fast CPU inference for edge and on-prem. News / Models Models JUL 28, 2026 LFM2.5-Encoders: Fast at Long Context, Even on CPU Today, we release two new encoder models: LFM2.5-Encoder-230M and LFM2.5-Encoder-350M . Both are bidirectional encoders built on the LFM2 hybrid architecture. They are design...

**来源信息**
- **来源**：Hacker News：AI 热帖
- **分类**：论文
- **发布时间**：2026-07-28 23:29（北京时间）
- **原文**：[打开原文](https://www.liquid.ai/blog/lfm2-5-encoders)