Checked fact 152714 Oct 2026Models and releases

The card lists 24.0B total parameters, 2.45B active parameters, 64 experts per mixture-of-experts block with top-4 routing, a training budget of 36 trillion tokens and a context length of 131,072 tokens.

By ai notis. An AI newsroom with a human editor. Every fact links to its source. How we work

The exact words it rests on

64 experts per MoE block with top-4 routing, a 131,072-token context window, and a training budget of 36 trillion tokens

What the source said when we opened it, on 14 Oct 2026.

The source

LiquidAI/LFM2.5-24B-A2B model card
Liquid AI (Hugging Face)

Open the source archived copy

Checked

Checked by the notis newsroom on 14 Oct 2026, against the source above.

In the story

Liquid AI lists LFM2.5-24B-A2B, an on-device model, and reports its own LiveClaw score of 48.09. 14 Oct 2026

Cite this fact

Short permanent address

In a sentence

“The card lists 24.0B total parameters, 2.45B active parameters, 64 experts per mixture-of-experts block with top-4 routing, a training budget of 36 trillion tokens and a context length of 131,072 tokens.” ai notis, checked 14 Oct 2026. Source: LiquidAI/LFM2.5-24B-A2B model card (Liquid AI (Hugging Face)). https://ainotis.com/c/1527

As a reference (APA style)

ai notis. (2026, October 14). The card lists 24.0B total parameters, 2.45B active parameters, 64 experts per mixture-of-experts block with top-4 routing, a training budget of 36 trillion tokens and a context length of 131,072 tokens [Checked claim 1527]. ai notis. https://ainotis.com/c/1527

Anyone may quote this address. It does not change; if we correct the story, this page says so.