by Liquid · Released 2026-02-25
LFM2-24B-A2B is the largest model in the LFM2 family of hybrid architectures designed for efficient on-device deployment. Built as a 24B parameter Mixture-of-Experts model with only 2B active parameters per...
What this model can take in, what it replies with, and what it does not do.
Standard OpenAI fields, grouped by what they control. Anything not listed is accepted and ignored.
The hard numbers — what it can take in, what it can give back, and where it stops.
No. Liquid has retired LFM2-24B-A2B and it can no longer be called through OpenDunes. This page is kept as a record of what it did. The rest of the catalog is unaffected — every model still listed is billed per token in dinars from the same balance.
LFM2-24B-A2B reads up to 128K tokens in one request.
It reads text and replies with text. It also supports replying in structured JSON and streaming its answer as it writes.
Yes. Top up your OpenDunes balance in dinars with a CIB or Edahabia card, then spend it on any model in the catalog from that same balance. No foreign card, no currency conversion, no subscription. LFM2-24B-A2B is not among them any more, but its replacements from the same lab are.
LFM2-24B-A2B is built by Liquid, released on Feb 25, 2026. OpenDunes served it unchanged — same model, same weights, billed in dinars — until the lab retired it.