LFM2-VL 3B

Parameters

3B

Context Length

33K

Modality

Text

Architecture

Dense

KV Cache Rate

60 KB/tok

License

other

Release Date

5 Oct 2026

Knowledge Cutoff

-

API Pricing (per 1M)

Self-hosted only

Hardware & System Requirements

Precision

Native

Context

Workload Memory: 4.37 GB

INT4 · 8K context

Base Weights: 1.40 GB

KV Cache: 0.47 GB

Runtime & Graphs: 2.50 GB

Minimum Required Hardware

1x NVIDIA L4 24GB

4.4 / 21.6 GB usable

1x RTX 3060 12GB

4.4 / 10.8 GB usable

Apple M4 24GB

4.4 / 18.0 GB usable

Determine requirements for your specific workload:

Open in VRAM Calculator

Runtime requirements

Checkpoint format

BF16

CUDA

>= 11.8

PyTorch

>= 2.1

Transformers

>= 5.14.1

ArchitecturePrecisionAttention kernel

Blackwell, Hopper

Native BF16

FlashAttention-3

Ada Lovelace, Ampere

Native BF16

FlashAttention-2

Architecture Diagram

Evaluation Benchmarks

No evaluation benchmarks for LFM2-VL 3B available.

Rankings

Overall Rank

-

Coding Rank

-

About LFM2-VL 3B

LFM2-VL 3B is a compact multimodal language foundation model developed by Liquid AI based on the Liquid Foundation Model architecture. It is optimized for efficient on-device visual reasoning, document parsing, and edge deployment.

Specifications

Attention

Attention Structure

Grouped-Query Attention

Attention Heads

32

Key-Value Heads

8

Attention Head Dimension

64

Position Embedding

ROPE

RoPE Theta

1,000,000

Linear Attention

Yes

Linear Attention Ratio

73.3%

Normalization

RMS Normalization

Activation Function

SwigLU

Dimensions

Base Precision

BF16

Hidden Dimension Size

2,048

Number of Layers

30

FFN Intermediate Size (Dense)

10,752

Tokenizer

Vocabulary Size

128,000

About LFM2

The LFM2 model family developed by Liquid AI.


Other LFM2 Models