liquid.ai3 weeks agoLFM2.5-24B-A2B: Built for Fast On-Device Agents | BlogLFM2.5-24B-A2B: an open-weight 24B reasoning model with 2B active parameters, built for fast, private on-device agents. Beats Qwen3.6-35B-A3B on LiveClaw.Visit liquid.aiBookmarkAdd to collection