Key Info
Liquid AI's LFM2.5-24B-A2B, the largest member of the LFM2.5 family, is now available on OpenRouter. It is a 24B-parameter MoE model with 2B active parameters per token, designed to reason before answering and built for agents and tool use.
Highlights
- 24B total parameters with 2B active per token (MoE architecture)
- Context window expanded to 131K tokens
- 128K-token vocabulary
- Positioned as an efficient local model for laptops and PCs handling multi-step tasks in popular agent harnesses
- Available to try now on OpenRouter