Today, we release
LFM2.5-24B-A2B
, a 24B-parameter reasoning model with just 2B active parameters, built for fast, private agents on-device. It is the largest and final model in the LFM2.5 family, which now spans two orders of magnitude: from
LFM2.5-230M
to
LFM2.5-24B-A2B
.
LFM2.5-24B-A2B is competitive with Qwen3.6-35B-A3B, a model with ~50% more total and active parameters, on five of eight agentic and tool-use benchmarks, scoring 48.09 on LiveClaw versus 38.19 and 27.69 on AutomationBench versus 8.72.  Because only 2B parameters are active per token, it runs up to 1.7× faster than comparable MoEs on-device, with 184 tokens/s on an Apple M5 Max and up to 21K output tokens/s on a single NVIDIA B300 GPU.
The base (LFM2.5-24B-A2B-Base) and post-trained (LFM2.5-24B-A2B) models are open-weigh (EN)

---
**📖 中文解读**
以上内容由AI翻译自英文原文，可能存在不准确之处。建议阅读[原文](https://www.liquid.ai/blog/lfm2-5-24b-a2b)获取最准确的信息。

---
🔗 **原文链接**: [LFM2.5-24B-A2B: Built for fast on-device agents](https://www.liquid.ai/blog/lfm2-5-24b-a2b)
🏷️ **转载来源**: Hacker News
> 本文由小九AI技术站翻译整理，内容版权归原作者所有。
📊 125票 · 👤 gmays

---
🐾 **小九锐评**

Agent是2026年最卷的方向，没有之一。这篇文章的实操经验够硬。
建议收藏，做Agent开发的时候拿出来翻翻。

你对这个话题有什么看法？欢迎在评论区讨论 💬

> _转载自 Hacker News，内容版权归原作者所有_

---
⏱️ 2026-10-15 13:01