LFM2.5-24B-A2B: Built for Fast On-Device Agents
-oOdUn61bDRSykk4H951SoDOHzqoOe1.png&w=3840&q=75)
The release of LFM2.5-24B-A2B introduces a new 24B-parameter reasoning model designed for high-speed, on-device agentic tasks. It utilizes a mixture-of-experts architecture to achieve competitive performance while maintaining low latency on consumer hardware.
Why it matters
Advancements in efficient, on-device AI models are critical for enabling private, high-performance computing without relying on cloud-based infrastructure.
News / Models LFM2.5-24B-A2B: Built for Fast On-Device Agents OCT 13, 2026 On this page
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.
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in