API Try Docs Blog

LFM3-VL-2B

LFM3-VL-2B is a vision-language model for on-device visual understanding. It takes text, images, and video as input, and combines the LFM3 language model with a new in-house vision encoder.

New in LFM3-VL-2B:

• Video understanding: the first LFM vision model that takes video as input. Ask questions about a recorded lecture, a walkthrough filmed on your phone, or a clip from a camera feed.
• Thinking: the model reasons step by step before it answers. This helps most with harder visual questions, such as math, charts, and multi-step problems.
• Spatial and embodied reasoning: the model finds objects and reasons about how they are placed relative to each other, a strong foundation for robotics applications.

Find more information about the LFM3 family in our release blog post.

Model details

---

The LFM3 family includes base, text, vision, audio, omni, and decision models.

LFM3-VL-2B is designed for on-device visual tasks: questions about video, reading documents, charts, and app screens, finding objects in an image, and spatial reasoning for robots. It is not recommended for coding, multi-turn conversations, or complex reasoning.

Inference

---

Use these ONNX weights with Transformers.js v4.3.2 or later. The example below uses AutoModelForImageTextToText and AutoProcessor with WebGPU.

npm install @huggingface/transformers@^4.3.2

import { AutoModelForImageTextToText, AutoProcessor, load_image } from "@huggingface/transformers";

const modelId = "onnx-community/LFM3-VL-2B-ONNX";
const model = await AutoModelForImageTextToText.from_pretrained(modelId, {
  device: "webgpu",
  dtype: {
    vision_encoder: "fp16",
    embed_tokens: "fp16",
    decoder_model_merged: "q4f16",
  },
});
const processor = await AutoProcessor.from_pretrained(modelId);

const image = await load_image(
  "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/cats.png",
);
const messages = [
  {
    role: "user",
    content: [
      { type: "image" },
      { type: "text", text: "What is in this image?" },
    ],
  },
];

// Format the chat and preprocess the image.
const chatPrompt = processor.apply_chat_template(messages, {
  add_generation_prompt: true,
  enable_thinking: false,
});
const inputs = await processor(image, chatPrompt, { add_special_tokens: false });

const outputs = await model.generate({
  ...inputs,
  do_sample: true,
  temperature: 0.1,
  top_k: 50,
  repetition_penalty: 1.1,
  max_new_tokens: 128,
});

// Decode only the newly generated tokens.
const inputLength = inputs.input_ids.dims.at(-1);
const generated = outputs.slice(null, [inputLength, null]);
console.log(processor.batch_decode(generated, { skip_special_tokens: true })[0]);

// Release the model when you are finished using it.
await model.dispose();

The processor loads the image preprocessing settings and chat template automatically. This example disables thinking with enable_thinking: false. Set it to true in processor.apply_chat_template() and increase max_new_tokens when you want to allow a longer reasoning trace.

Fine-tuning

---

LFM3-VL-2B is designed for easy customization for a given domain, task, or application via TRL.

Performance

---

Benchmarks

Across 62 vision benchmarks, LFM3-VL-2B scores at least 8 points ahead of all models of similar size. It matches Gemma 4 12B, is within 2 points of GPT-5.6 Luna, and scores higher than Gemini 3.5 Flash Lite.

All 62 benchmarks Evaluation details

We evaluated all models with vLLM 0.27.1, using each model's recommended generation parameters when available.

We used reasoning mode for all models except LFM2.5-VL-3B, which does not support it, and the LFM3-VL-2B no-thinking configuration, for which we disabled thinking. For InternVL3.5, we report the higher score from reasoning and non-reasoning modes on each benchmark. GPT-5.6 Luna and Gemini 3.5 Flash Lite used their default reasoning effort: medium for Luna and minimal for Gemini.

For models without native video support, we sampled frames at 2 FPS, up to a maximum of 384 frames per video, and provided them to the model as images.

All scores are normalized to a 0–100 scale. Each category score is the unweighted mean of its benchmark scores, and the Average row is the unweighted mean across all 62 benchmarks.

Original score scales: MME uses a 0–2800 scale; OCRBench, 0–1000; and DocVQA and InfographicVQA, 0–1.
RefCOCO-avg: Mean P@1 across eight splits: RefCOCO (val, testA, testB), RefCOCO+ (val, testA, testB), and RefCOCOg (val, test).
POPE: F1 over all examples from the adversarial, popular, and random subsets.
MMMU-Pro: The multiple-choice variant with 10 answer options per question.

Edge inference

LFM3-VL-2B has the highest prefill throughput and the lowest memory use of all compared models on all three devices.

Evaluation details

All models run in llama.cpp with a Q4_0 GGUF language model and an F16 vision projector (mmproj). Each request has one 512×512 image and 1,024 text tokens, and generates 256 tokens. Prefill is the prompt processing speed and decode is the generation speed, both in tokens per second. Memory is peak use in GB.

GPU inference

On one NVIDIA B300, LFM3-VL-2B produces more than 12,000 tokens/s at 1,024 concurrent requests, more than any compared model.

Evaluation details

All models run in BF16 on SGLang v0.5.22 with LFM3 support. Each request has one 512×512 image and 1,024 text tokens, and generates at most 256 tokens. Throughput is end-to-end output tokens per second, as the mean of 3 runs.

Training

---

Pre-training: LFM3-VL-2B starts from LFM3-1.7B-Base, which is pre-trained in two stages on 22T tokens in total.

• Main pre-training: compared with LFM2, the mix adds more scientific papers, repository-level code, math, and a wider range of knowledge sources.
• Mid-training: the last 3.5T tokens use curated, higher-quality data with a larger share of agentic traces and tool calls. This stage also extends the context window to 256K tokens.

Post-training: four stages develop visual understanding, reasoning, and tool use.

• Vision mid-training: LFM3-1.7B-Base is joined with our new in-house vision encoder, which combines convolution and attention, and trained on image and video data.
• Supervised fine-tuning (SFT): a wide mix of instruction following, reasoning, tool calling, agentic traces, grounding, and multimodal perception and understanding.
• Vision reinforcement learning: many domains trained at once, to improve reasoning and general intelligence.
• Multi-domain on-policy distillation (MOPD): a vision teacher model and the Mixed RL model from LFM3-1.7B are distilled back into a single model.

Citation

---

@article{liquidAI2026LFM3,
  author  = {Liquid AI},
  title   = {Introducing LFM3: Every Modality for Specialized Intelligence},
  journal = {Liquid AI Blog},
  year    = {2026},
  note    = {www.liquid.ai/blog/lfm3},
}

For questions and feedback, join Discord. For custom deployment and fine-tuning support, contact Liquid AI.