README.md

---
library_name: transformers
license: other
license_name: lfm1.0
license_link: LICENSE
language:
- en
- ja
- ko
- fr
- es
- de
- ar
- zh
pipeline_tag: image-text-to-text
tags:
- liquid
- lfm2
- lfm2-vl
- edge
- lfm2.5-vl
- lfm2.5
base_model: LiquidAI/LFM2.5-1.2B-Base
---

<center>
<div style="text-align: center;">
  <img 
    src="https://cdn-uploads.huggingface.co/production/uploads/61b8e2ba285851687028d395/2b08LKpev0DNEk6DlnWkY.png" 
    alt="Liquid AI"
    style="width: 100%; max-width: 100%; height: auto; display: inline-block; margin-bottom: 0.5em; margin-top: 0.5em;"
  />
</div>
<div style="display: flex; justify-content: center; gap: 0.5em;">
<a href="https://playground.liquid.ai/chat?model=lfm2.5-vl-1.6b"><strong>Try LFM</strong></a> • <a href="https://docs.liquid.ai/lfm/getting-started/intro"><strong>Documentation</strong></a> • <a href="https://leap.liquid.ai/"><strong>LEAP</strong></a> • <a href="https://huggingface.co/spaces/LiquidAI/LFM2.5-VL-1.6B-WebGPU"><strong>WebGPU demo</strong></a></a> 
</div>
</center>

# LFM2.5‑VL-1.6B

LFM2.5‑VL-1.6B is [Liquid AI](https://www.liquid.ai/)'s refreshed version of the first vision-language model, [LFM2-VL-1.6B](https://huggingface.co/LiquidAI/LFM2-VL-1.6B), built on an updated backbone [LFM2.5-1.2B-Base](https://huggingface.co/LiquidAI/LFM2.5-1.2B-Base) and tuned for stronger real-world performance. Find more about LFM2.5 family of models in our [blog post](https://www.liquid.ai/blog/introducing-lfm2-5-the-next-generation-of-on-device-ai).

* **Enhanced instruction following** on vision and language tasks.
* **Improved multilingual vision understanding** in Arabic, Chinese, French, German, Japanese, Korean, and Spanish.
* **Robust understanding of visual content** with improved results on multi-image inputs, high-resolution images, and OCR.

🎥⚡️ You can try LFM2.5-VL-1.6B running locally in your browser with our real-time video stream captioning [WebGPU demo](https://huggingface.co/spaces/LiquidAI/LFM2.5-VL-1.6B-WebGPU) 🎥⚡️ 

Alternatively, try the API model on the [Playground](https://playground.liquid.ai/chat?model=lfm2.5-vl-1.6b).


## 📄 Model details

| Model | Parameters | Description |
|-------|------------|-------------|
| [LFM2.5-1.2B-Base](https://huggingface.co/LiquidAI/LFM2.5-1.2B-Base) | 1.2B | Pre-trained base model for fine-tuning |
| [LFM2.5-1.2B-Instruct](https://huggingface.co/LiquidAI/LFM2.5-1.2B-Instruct) | 1.2B | General-purpose instruction-tuned model |
| [LFM2.5-1.2B-JP](https://huggingface.co/LiquidAI/LFM2.5-1.2B-JP) | 1.2B | Japanese-optimized chat model |
| [**LFM2.5-VL-1.6B**](https://huggingface.co/LiquidAI/LFM2.5-VL-1.6B) | 1.6B | Vision-language model with fast inference |
| [LFM2.5-Audio-1.5B](https://huggingface.co/LiquidAI/LFM2.5-Audio-1.5B) | 1.5B | Audio-language model for speech and text I/O |

LFM2.5-VL-1.6B is a general-purpose vision-language model with the following features:

- **LM Backbone**: LFM2.5-1.2B-Base
- **Vision encoder**: SigLIP2 NaFlex shape‑optimized 400M
- **Context length**: 32,768 tokens
- **Vocabulary size**: 65,536
- **Languages**: English, Arabic, Chinese, French, German, Japanese, Korean, and Spanish
- **Native resolution processing**: handles images up to 512*512 pixels without upscaling and preserves non-standard aspect ratios without distortion
- **Tiling strategy**: splits large images into non-overlapping 512×512 patches and includes thumbnail encoding for global context
- **Inference-time flexibility**: user-tunable maximum image tokens and tile count for speed/quality tradeoff without retraining
- **Generation parameters**: 
  - text: `temperature=0.1`, `min_p=0.15`, `repetition_penalty=1.05`
  - vision: `min_image_tokens=64` `max_image_tokens=256`, `do_image_splitting=True`

| Model | Description |
|-------|-------------|
| [**LFM2.5-VL-1.6B**](https://huggingface.co/LiquidAI/LFM2.5-VL-1.6B) | Original model checkpoint in native format. Best for fine-tuning or inference with Transformers and vLLM. |
| [LFM2.5-VL-1.6B-GGUF](https://huggingface.co/LiquidAI/LFM2.5-VL-1.6B-GGUF) | Quantized format for llama.cpp and compatible tools. Optimized for CPU inference and local deployment with reduced memory usage. |
| [LFM2.5-VL-1.6B-ONNX](https://huggingface.co/LiquidAI/LFM2.5-VL-1.6B-ONNX) | ONNX Runtime format for cross-platform deployment. Enables hardware-accelerated inference across diverse environments (cloud, edge, mobile). |
| [LFM2.5-VL-1.6B-MLX](https://huggingface.co/mlx-community/LFM2.5-VL-1.6B-8bit) | MLX format for Apple Silicon. Optimized for fast inference on Mac devices using the MLX framework. |

We recommend using it for general vision-language workloads, OCR or document comprehension. It’s not well-suited for knowledge-intensive tasks.

### Chat Template

LFM2.5-VL uses a ChatML-like format. See the [Chat Template documentation](https://docs.liquid.ai/lfm/getting-started/vision#chat-template) for details.

```
<|startoftext|><|im_start|>system
You are a helpful multimodal assistant by Liquid AI.<|im_end|>
<|im_start|>user
<image>Describe this image.<|im_end|>
<|im_start|>assistant
This image shows a Caenorhabditis elegans (C. elegans) nematode.<|im_end|>
```

You can use [`processor.apply_chat_template()`](https://huggingface.co/docs/transformers/en/chat_templating_multimodal) to format your messages automatically.

## 🏃 Inference

You can run LFM2.5-VL-1.6B with Hugging Face [`transformers`](https://github.com/huggingface/transformers):

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

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

# Load model and processor
model_id = "LiquidAI/LFM2.5-VL-1.6B"
model = AutoModelForImageTextToText.from_pretrained(
    model_id,
    device_map="auto",
    dtype="bfloat16"
)
processor = AutoProcessor.from_pretrained(model_id)

# Load image and create conversation
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?"},
        ],
    },
]

# Generate Answer
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=64)
processor.batch_decode(outputs, skip_special_tokens=True)[0]

# This image showcases the iconic Statue of Liberty standing majestically on Liberty Island in New York Harbor. The statue is positioned on a small island surrounded by calm blue waters, with the New York City skyline visible in the background.
```

### Tool Use

LFM2.5 supports function calling for text only input by applying the chat template with the tokenizer. See the [Tool Use documentation](https://docs.liquid.ai/lfm/key-concepts/tool-use) for the full guide.

```python
tools = [{
    "name": "get_weather",
    "description": "Get current weather for a location",
    "parameters": {
        "type": "object",
        "properties": {"location": {"type": "string"}},
        "required": ["location"]
    }
}]

messages = [{"role": "user", "content": "What's the weather in Paris?"}]

# Apply chat template with tools
inputs = processor.tokenizer.apply_chat_template(
    messages,
    tools=tools,
    add_generation_prompt=True,
    return_tensors="pt",
    return_dict=True,
)
input_ids = inputs["input_ids"].to(model.device)
outputs = model.generate(input_ids, max_new_tokens=256)
response = processor.tokenizer.decode(outputs[0, input_ids.shape[1]:], skip_special_tokens=False)

# <|tool_call_start|>[get_weather(location="Paris")]<|tool_call_end|>I am retrieving the current weather for Paris.<|im_end|>
```

| Name | Description | Docs | Notebook |
|------|-------------|------|----------|
| [Transformers](https://github.com/huggingface/transformers) | Simple inference with direct access to model internals. | <a href="https://docs.liquid.ai/lfm/inference/transformers#vision-models">Link</a>| <a href="https://colab.research.google.com/drive/1WVQpf4XrHgHFkP0FnlZfx2nK8PugvQNZ?usp=sharing"><img src="https://cdn-uploads.huggingface.co/production/uploads/61b8e2ba285851687028d395/vlOyMEjwHa_b_LXysEu2E.png" width="110" alt="Colab link"></a> |
| [vLLM](https://github.com/vllm-project/vllm) | High-throughput production deployments with GPU. | coming soon | <a href="https://colab.research.google.com/drive/1sUfQlqAvuAVB4bZ6akYVQPGmHtTDUNpF?usp=sharing"><img src="https://cdn-uploads.huggingface.co/production/uploads/61b8e2ba285851687028d395/vlOyMEjwHa_b_LXysEu2E.png" width="110" alt="Colab link"></a> |
| [llama.cpp](https://github.com/ggml-org/llama.cpp) | Cross-platform inference with CPU offloading. | <a href="https://docs.liquid.ai/lfm/inference/llama-cpp#vision-models">Link</a> | <a href="https://colab.research.google.com/drive/1q2PjE6O_AahakRlkTNJGYL32MsdUcj7b?usp=sharing"><img src="https://cdn-uploads.huggingface.co/production/uploads/61b8e2ba285851687028d395/vlOyMEjwHa_b_LXysEu2E.png" width="110" alt="Colab link"></a> |

## 🔧 Fine-tuning

We recommend fine-tuning LFM2.5-VL-1.6B model on your use cases to maximize performance.

| Notebook  | Description                                                          | Link |
|-----------|----------------------------------------------------------------------|------|
| SFT (TRL) | Supervised Fine-Tuning with LoRA using TRL. | <a href="https://colab.research.google.com/drive/10530_jt_Joa5zH2wgYlyXosypq1R7PIz?usp=sharing"><img src="https://cdn-uploads.huggingface.co/production/uploads/61b8e2ba285851687028d395/vlOyMEjwHa_b_LXysEu2E.png" width="110" alt="Colab link"></a> |


## 📊 Performance

| Model              | MMStar | MM-IFEval | BLINK | InfoVQA (Val) | OCRBench (v2) | RealWorldQA | MMMU (Val) | MMMB (avg) | Multilingual MMBench (avg) |
|--------------------|--------|-----------|-------|---------------|---------------|-------------|------------|------------|----------------------------|
| **LFM2.5-VL-1.6B** | 50.67  | 52.29     | 48.82 | 62.71         | 41.44         | 64.84       | 40.56      | 76.96      | 65.90                      |
| LFM2-VL-1.6B       | 49.87  | 46.35     | 44.50 | 58.35         | 35.11         | 65.75       | 39.67      | 72.13      | 60.57                      |
| InternVL3.5-1B     | 50.27  | 36.17     | 44.19 | 60.99         | 33.53         | 57.12       | 41.89      | 68.93      | 58.32                      |
| FastVLM-1.5B       | 53.13  | 24.99     | 43.29 | 23.92         | 26.61         | 61.56       | 38.78      | 64.84      | 50.89                      |

All vision benchmark scores are obtained using [VLMEvalKit](https://github.com/open-compass/VLMEvalKit). Multilingual scores are based on the average of benchmarks translated by GPT-4.1-mini from English to Arabic, Chinese, French, German, Japanese, Korean, and Spanish.

## 📬 Contact

If you are interested in custom solutions with edge deployment, please contact [our sales team](https://www.liquid.ai/contact).

## Citation

```
@article{liquidai2025lfm2,
 title={LFM2 Technical Report},
 author={Liquid AI},
 journal={arXiv preprint arXiv:2511.23404},
 year={2025}
}
```a