Liquid AI
1 story · first 14 Oct · latest 14 Oct
Products and models
Facts we checked
Each fact was checked against the page it links to.
-
The card reports, as Liquid AI's own figures, a score of 48.09 on LiveClaw against 38.19 for Qwen3.6-35B-A3B.
LiquidAI/LFM2.5-24B-A2B model card · Liquid AI (Hugging Face) · archived copy · checked 14 Oct 2026 · Cite this fact · the story
-
Liquid AI's model card on Hugging Face describes LFM2.5-24B-A2B as a 24B-parameter reasoning model with only 2B active parameters, built for fast, private agents on-device.
LiquidAI/LFM2.5-24B-A2B model card · Liquid AI (Hugging Face) · archived copy · checked 14 Oct 2026 · Cite this fact · the story
The story
Built from our published stories and the facts we checked, and kept up to date by them. How we correct.