LFM2-24B-A2B

LLM

Liquid AI

LFM2-24B-A2B is a text chat model built for efficient inference, featuring 24B total parameters with only 2B active at a time via a mixture-of-experts design, offering a 32,768-token context window.

Context tokens

32,768

Output tokens

8,192

Docs

Model documentation

Released

Feb 24, 2026

Schema

This model is not optimized for coding tasks. It responds without reasoning traces and has a low default temperature (0.1), so outputs may be more deterministic than other models.

Schema documentation

Capabilities

Tool use

Structured output

Multilingual

Supported languages

Supported tools

No tools enabled.

Pricing

Type	Credits	Units
Input	0.40	Credits per 1k tokens
Output	1.60	Credits per 1k tokens

Variants