LFM2-24B-A2B

Name: LFM2-24B-A2B
Brand: LiquidAI

24B

by LiquidAI

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 token, it delivers high-quality generation while maintaining low inference costs. The model fits within 32 GB of RAM, making it practical to run on consumer laptops and desktops without sacrificing capability.

Chat with LFM2-24B-A2B

Input Price$0.00/1M tokens

Output Price$0.00/1M tokens

Intelligence7.9

Coding1.4

Specifications

Technical details and pricing.

ProviderLiquidAI

Context Window32,768 tokens

Release DateSep 23, 2025

ModalitiesText

Benchmarks

10 benchmark scores from Artificial Analysis.

GPQA30.6%

MMLU Pro29.8%

HLE5.2%

LiveCodeBench8.1%

AIME 20258.3%

SciCode2.5%

LCR0.0%

IFBench19.5%

Tau213.5%

TerminalBench Hard0.8%

Composite Indices

Intelligence, Coding, Math

Standard Benchmarks

Academic and industry benchmarks

Frequently Asked Questions

What is LFM2-24B-A2B good for?

Use LFM2-24B-A2B for everyday tasks like writing, summarizing, brainstorming, and getting clear explanations.

How much does LFM2-24B-A2B cost?

Pricing is based on usage. Current rates are $0.00/1M tokens for input and $0.00/1M tokens for output.

Can I try LFM2-24B-A2B for free?

Yes. You can start a chat instantly and test the model before deciding on a plan.

Does LFM2-24B-A2B support images or audio?

LFM2-24B-A2B focuses on text-based tasks.

Similar Models

Other models you might want to explore.

LFM2-2.6B

LiquidAI

LFM2 is a new generation of hybrid models developed by Liquid AI, specifically designed for edge AI and on-device deployment.

Details →

LFM2-8B-A1B

LiquidAI

LFM2-8B-A1B is an efficient on-device Mixture-of-Experts (MoE) model from Liquid AI’s LFM2 family, built for fast, high-quality inference on edge hardware.

Details →

Mistral Small 3.2 24B

Mistral

Mistral-Small-3.2-24B-Instruct-2506 is an updated 24B parameter model from Mistral optimized for instruction following, repetition reduction, and improved function calling.

Details →

Benchmarks and pricing are sourced from Artificial Analysis where available. OpenRouter specs are used as a fallback.