# Liquid AI releases open-weight reasoning model LFM2.5-24B-A2B



Liquid AI



Liquid AI released LFM2.5-24B-A2B, a reasoning MoE model for agents and tool use, with 24B total parameters and 2B active per token. Base and post-trained weights are available.


## Details

Liquid AI released LFM2.5-24B-A2B as the largest member of the LFM2.5 family, intended for multi-step agent tasks on laptops and PCs. It has 24B total parameters and 2B active per token, with a 131K context stated by Liquid AI and open base and post-trained weights. OpenRouter also offers the model: its catalog lists a 128,000-token context, text input and pricing of $0.10 per million input tokens and $1 per million output tokens.

## Availability

Base and post-trained weights are available from Liquid AI; OpenRouter access is also available at the platform's listed rates.
- Released: Oct 13, 2026
- Review updated: Nov 05, 2026 00:43 Thu · Beijing Time (UTC+8)

## Sources

- Source excerpt: https://x.com/liquidai/status/2110055807194931348
- Source excerpt: https://x.com/OpenRouter/status/2110068048749355303
- Source excerpt: https://openrouter.ai/api/v1/models

Canonical entry: https://www.ailyre.com/source/?id=liquid-ai-lfm2-5-24b-a2b&lang=en