Sobre

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. It uses 8.3B total parameters with only ~1.5B active per token, delivering strong performance while keeping compute and memory usage low—making it ideal for phones, tablets, and laptops.

Família do Modelo

LFM2 24B A2B2026-02-25 LiquidAI: LFM2-24B-A2B2026-02-25 LFM2.5-1.2B-Thinking2026-01-20 LiquidAI: LFM2.5-1.2B-Instruct (free)2026-01-20 LiquidAI: LFM2.5-1.2B-Thinking (free)2026-01-20 LFM2.5-1.2B-Instruct2026-01-05 LFM2.5-VL-1.6B2026-01-05 LiquidAI: LFM2-2.6B2025-10-20

Benchmarks

MMLU-Pro

50.5%

GPQA Diamond

34.4%

HLE

4.9%

LiveCodeBench

15.1%

SciCode

6.8%

TerminalBench Hard

0.0%

MATH-500Não avaliado

AIMENão avaliado

AIME 2025

25.3%

IFBench

26.3%

Long Context Recall

0.0%

Tau2

10.5%

Média do MercadoMelhor Score

LFM2 8B A1B

Sobre

Família do Modelo

Posição no Mercado

Preços

Calculadora de Custo

vs. Modelos Similares

Desempenho

Benchmarks

Comparação Rápida

Modelos Similares