> ## Documentation Index > Fetch the complete documentation index at: https://docs.liquid.ai/llms.txt > Use this file to discover all available pages before exploring further. # LFM2.5-VL-450M > Compact 450M vision-language model with enhanced performance for edge deployment export const VlLlamacpp = ({ggufRepo, samplingFlags}) =>

llama.cpp enables efficient CPU inference for vision models.

Install:


{`brew install llama.cpp`.split('\n').map((line, i) => {line}{'\n'})}

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

Run:


{`llama-cli \\
    -hf ${ggufRepo}:Q4_0 \\
    --image test_image.jpg \\
    -p "What's in this image?" \\
    -n 128 \\
    ${samplingFlags}`.split('\n').map((line, i) => {line}{'\n'})}

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.

; export const VlSglang = ({modelId}) =>

Install:


{`uv pip install "sglang>=0.5.10"`.split('\n').map((line, i) => {line}{'\n'})}

Launch server:


{`sglang serve \\
    --model-path ${modelId} \\
    --host 0.0.0.0 \\
    --port 30000`.split('\n').map((line, i) => {line}{'\n'})}

Query (OpenAI-compatible):


{`from openai import OpenAI

client = OpenAI(base_url="http://localhost:30000/v1", api_key="None")

response = client.chat.completions.create(
    model="${modelId}",
    messages=[{
        "role": "user",
        "content": [
            {"type": "image_url", "image_url": {"url": "http://images.cocodataset.org/val2017/000000039769.jpg"}},
            {"type": "text", "text": "Describe what you see in this image."},
        ],
    }],
    temperature=0.0,
    max_tokens=256,
)

print(response.choices[0].message.content)`.split('\n').map((line, i) => {line}{'\n'})}
; export const VlVllm = ({modelId}) =>

Install:


{`uv pip install vllm==0.19.0`.split('\n').map((line, i) => {line}{'\n'})}



{`uv pip install transformers==5.5.0 pillow`.split('\n').map((line, i) => {line}{'\n'})}

Run:


{`from vllm import LLM, SamplingParams

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

llm = LLM(
    model="${modelId}",
    max_model_len=1024,
)

sampling_params = SamplingParams(
    temperature=0.1,
    min_p=0.15,
    repetition_penalty=1.05,
    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)`.split('\n').map((line, i) => {line}{'\n'})}
; export const VlTransformers = ({modelId}) =>

Install:


{`pip install "transformers>=5.2.0" pillow torch`.split('\n').map((line, i) => {line}{'\n'})}

Download & Run:


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

model_id = "${modelId}"
model = AutoModelForImageTextToText.from_pretrained(
    model_id,
    device_map="auto",
    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, do_sample=True, temperature=0.1, min_p=0.15, repetition_penalty=1.05, max_new_tokens=256)
response = processor.batch_decode(outputs, skip_special_tokens=True)[0]
print(response)`.split('\n').map((line, i) => {line}{'\n'})}
; ← Back to Vision Models LFM2.5-VL-450M is Liquid AI's smallest vision-language model, building on LFM2-VL-450M with extended reinforcement learning for improved performance while maintaining the same compact deployment footprint.
HF GGUF ONNX
## Specifications | Property | Value | | -------------- | ----------------- | | Parameters | 450M | | Context Length | 32K tokens | | Architecture | LFM2.5-VL (Dense) |
Minimal memory footprint Fastest vision model inference Runs on mobile and embedded devices
## Quick Start