LFM2-VL-450M

Compact 450M vision-language model for edge deployment and fast inference

← Back to Vision Models

LFM2-VL-450M is Liquid AI’s smallest vision-language model, designed for edge deployment with strict memory and compute constraints. Delivers fast multimodal inference on resource-limited devices.

HFGGUFMLXONNX

Specifications

Property	Value
Parameters	450M
Context Length	32K tokens
Architecture	LFM2-VL (Dense)

Ultra-Light

Minimal memory footprint

Low Latency

Fastest vision model inference

Edge-Ready

Runs on mobile and embedded devices

Quick Start

Transformers

Install:

pip install "transformers>=5.0.0" pillow torch

Transformers v5 is newly released. If you encounter issues, fall back to the pinned git source:

pip install git+https://github.com/huggingface/transformers.git@3c2517727ce28a30f5044e01663ee204deb1cdbe pillow torch

Download & Run:

from transformers import AutoProcessor, AutoModelForImageTextToText
from transformers.image_utils import load_image

model_id = "LiquidAI/LFM2-VL-450M"
model = AutoModelForImageTextToText.from_pretrained(
    model_id,
    device_map="auto",
    torch_dtype="bfloat16"
)
processor = AutoProcessor.from_pretrained(model_id)

url = "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
image = load_image(url)

conversation = [
    {
        "role": "user",
        "content": [
            {"type": "image", "image": image},
            {"type": "text", "text": "What is in this image?"},
        ],
    },
]

inputs = processor.apply_chat_template(
    conversation,
    add_generation_prompt=True,
    return_tensors="pt",
    return_dict=True,
    tokenize=True,
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=256)
response = processor.batch_decode(outputs, skip_special_tokens=True)[0]
print(response)

vLLM

vLLM support for LFM Vision Models requires a specific version. Install from the custom source below.

Install:

VLLM_PRECOMPILED_WHEEL_COMMIT=72506c98349d6bcd32b4e33eec7b5513453c1502 \
  VLLM_USE_PRECOMPILED=1 \
  pip install git+https://github.com/vllm-project/vllm.git

pip install "transformers>=5.0.0" pillow

Transformers v5 is newly released. If you encounter issues, fall back to the pinned git source:

pip install git+https://github.com/huggingface/transformers.git@3c2517727ce28a30f5044e01663ee204deb1cdbe pillow

Run:

from vllm import LLM, SamplingParams

IMAGE_URL = "http://images.cocodataset.org/val2017/000000039769.jpg"

llm = LLM(
    model="LiquidAI/LFM2-VL-450M",
    max_model_len=1024,
)

sampling_params = SamplingParams(
    temperature=0.0,
    max_tokens=256,
)

messages = [{
    "role": "user",
    "content": [
        {"type": "image_url", "image_url": {"url": IMAGE_URL}},
        {"type": "text", "text": "Describe what you see in this image."},
    ],
}]

outputs = llm.chat(messages, sampling_params)
print(outputs[0].outputs[0].text)

llama.cpp

llama.cpp enables efficient CPU inference for vision models.

Install:

brew install llama.cpp

Or download pre-built binaries from llama.cpp releases.

Run:

llama-cli \
    -hf LiquidAI/LFM2-VL-450M-GGUF:Q4_0 \
    --image test_image.jpg \
    -p "What's in this image?" \
    -n 128

The -hf flag downloads the model directly from Hugging Face. Use --image-max-tokens to control image token budget.

For server deployment and advanced usage, see the llama.cpp guide.