Liquid AI/Mixture of Experts

LFM2 8B A1B

Name: LFM2 8B A1B
Author: Liquid AI

LFM2 is a new generation of hybrid models developed by Liquid AI, specifically designed for edge AI and on-device deployment. It sets a new standard in terms of quality, speed, and memory efficiency.

chattool_usemultilingual

8.3B

Parameters (1.5B active)

125K

Context length

Benchmarks

Quantizations

Architecture

MoE

Released

2025-11-15

Layers

KV Heads

Head Dim

Family

lfm

Quantization Options

Quant	Bits	VRAM	Quality
Q4_K_M	4.89	5.6 GB	good
Q5_K_S	5.57	6.3 GB	good
Q5_K_M	5.7	6.4 GB	good
Q6_K	6.56	7.3 GB	excellent
Q8_0	8.5	9.3 GB	lossless
FP16	16	17.1 GB	lossless

Select your GPU above to see speed estimates and compatibility for each quantization.

▸ READY TO RUN THIS?RENT BY THE HOUR

RENT A GPU AND RUN LFM2 8B A1B NOW

Rent on RunPod →Or Vast.ai →

Spin up an A100 / H100 / 4090 in ~60s. Pay by the second. Cancel anytime.

Community Ratings

Loading ratings...

Benchmarks (14)

IFEval77.6

MMLU-PRO50.5

GPQA Diamond34.4

GPQA29.3

IFBench26.3

AIME25.3

AA Math25.3

MATH-50025.3

LiveCodeBench15.1

τ²-Bench10.5

AA Intelligence7.0

SciCode6.8

HLE4.9

AA Coding2.3

Run this model

▸Easiest way to get started·Beginners

DOCS ↗

curl -fsSL https://ollama.com/install.sh | sh

$ollama run lfm:8b-q4_K_M

Tag may need adjustment — check ollama.com/library/lfm for available tags.

▸ SETUP GUIDE

Auto-setup with fitmyllm CLI

Detects your GPU, recommends the best model, downloads it, and starts chatting — zero config. Benchmarks your speed and contributes anonymous data to improve predictions.

pip install fitmyllmthen run fitmyllmLearn more

Auto-detect GPULive tok/s in chatSpeed benchmarks9 inference engines

HuggingFace GGUF Downloads Build Hardware

GPUs that can run this model

At Q4_K_M quantization. Sorted by minimum VRAM.

NVIDIA RTX 3050 6GB

6 GB VRAM • 168 GB/s

NVIDIA

$169