Synaptics/LiquidAI-LFM2-VL-450M

LFM2-VL-450M — Torq build (Synaptics SL2619 NPU)

This repository provides compiled model files for LiquidAI's LFM2-VL-450M vision-language model, ready to run on the Synaptics SL2610-series Torq NPU. Give it an image and a natural-language question, and it answers questions about that image.

Quick start guide:

• Buy a Machina kit: Get an SL2600 Machina kit delivered to you
• Torq Examples: Use Torq-examples LiquidAI/LiquidAI-LFM2-VL-450M scripts to download and deploy on your Machina kit

Model Overview

LFM2‑VL is designed to process text and images with variable resolutions. Built on the LFM2 backbone, it is optimized for low-latency and edge AI applications.

LFM2-VL utilizes hybrid conv/attention text decoders that execute on the NPU in bf16; the token embeddings run on the host CPU.

Image + prompt → caption / visual question answering. The image is encoded once and its KV cache is reused, so follow-up questions about the same image stay fast.

Model Features

Contents

File	Size	Size (W8)	Role
vision_encoder_256.vmfb	203 MB	106 MB	SigLIP vision encoder, 256-res → 64 image tokens
decoder_image_2part_A.vmfb	353 MB	150 MB	one-shot image-prefill decoder, layers 0–7
decoder_image_2part_B.vmfb	311 MB	132 MB	one-shot image-prefill decoder, layers 8–15
decoder_nolm.vmfb	577 MB	290 MB	LFM2 single-token decode body (hidden-state output)
lm_head.vmfb	134 MB	67 MB	tied LM head (hidden → 65 536 logits)
token_embeddings.npy	134 MB	—	CPU embedding LUT / tied-LM-head weights (bf16)
config.json, tokenizer.json	—	—	model config + tokenizer
cats-and-dogs-256.jpg	—	—	sample 256-res image for the demo
onnx/	~2 GB	—	reference ONNX exports (vision encoder, merged decoder, embeddings) for non-Torq runtimes

Model Details

• Base model: LiquidAI LFM2-VL-450M (SigLIP vision tower + LFM2 language model).
• Text decoder: LFM2 — hidden size 1024, 16 layers, 16 attention heads, vocabulary 65 536, hybrid short-convolution + grouped-query attention.
• Image tokens: 64 per image (256-resolution input).
• Precision: bf16 on the NPU.
• Target: Synaptics SL2619, compiled with the Torq compiler.
• On-device performance (SL2619, indicative): vision encode ~2.4 s, image→KV prefill ~3.7 s, decode ~3.6–4.2 tok/s.

Tested Platforms

• Synaptics Astra™ Machina SL2619 2GB

Metrics

Platform	Model / Stage	Environment	NPU Clock	TTFT	Infer / s
SL2619 2GB	LFM2-VL-450M	Torq v2.0.0	1 GHz	2844 ms	3.4
SL2619 2GB	LFM2-VL-450M-W8	Torq v2.0.0	1 GHz	1371 ms	6.2

Deployment

The models have been tested with the following environment.

• Torq Compiler: v2.0.0
• Torq Runtime: v2.0.0 included in Astra SDK release scarthgap_6.12_v2.4.0

Usage Tutorials / Example Apps

A usage example is provided in the Torq Examples / LiquidAI-LFM2-VL-450M.

Check out the README for instructions.

License & attribution

This repository is a redistribution of a model created by Liquid AI, Inc., licensed under the LFM Open License v1.0. Copies of the license and the attribution notices are included alongside the model files:

• LICENSE — a verbatim copy of the LFM Open License v1.0.
• NOTICE — the copyright, patent, trademark, and attribution notices retained from the original Work (per Section 4(c) of the license).

Original model: LFM2.5-230M · Copyright © Liquid AI, Inc.

Learn More

• Synaptics AI Developer Zone: Get started with documentation, tutorials and resources for your Edge AI journey.
• Torq Compiler Documentation: Learn more about the Torq compiler based on MLIR and IREE.
• Synaptics Astra SDK: Learn more about the Yocto Project-based Linux software available for Astra SL processors.
• Astra Support Portal: Connect with our engineering team and community.