📃 License • 💻 Code • 📑 Technical Report • 📊 Benchmarks • 🚀 Getting Started • 🤗 Models

🎯 Introduction

Youtu-VL is a lightweight yet robust Vision-Language Model (VLM) built on the Youtu-LLM with 4B parameters. It pioneers Vision-Language Unified Autoregressive Supervision (VLUAS), which markedly strengthens visual perception and multimodal understanding. This enables a standard VLM to perform vision-centric tasks without task-specific additions. Across benchmarks, Youtu-VL stands out for its versatility, achieving competitive results on both vision-centric and general multimodal tasks.

✨ Key Features

• Comprehensive Vision-Centric Capabilities: The model demonstrates strong, broad proficiency across classic vision-centric tasks, delivering competitive performance in visual grounding, image classification, object detection, referring segmentation, semantic segmentation, depth estimation, object counting, and human pose estimation.

• Promising Performance with High Efficiency: Despite its compact 4B-parameter architecture, the model achieves competitive results across a wide range of general multimodal tasks, including general visual question answering (VQA), multimodal reasoning and mathematics, optical character recognition (OCR), multi-image and real-world understanding, hallucination evaluation, and GUI agent tasks.

🤗 Model Download

Model Name	Description	Download
Youtu-VL-4B-Instruct	Visual language model of Youtu-LLM	🤗 Model
Youtu-VL-4B-Instruct-GGUF	Visual language model of Youtu-LLM, in GGUF format	🤗 Model

🧠 Model Architecture Highlights

• Vision–Language Unified Autoregressive Supervision (VLUAS): Youtu-VL is built on the VLUAS paradigm to mitigate the text-dominant optimization bias in conventional VLMs, where visual signals are treated as passive conditions and fine-grained details are often dropped. Rather than using vision features only as inputs, Youtu-VL expands the text lexicon into a unified multimodal vocabulary through a learned visual codebook, turning visual signals into autoregressive supervision targets. Jointly reconstructing visual tokens and text explicitly preserves dense visual information while strengthening multimodal semantic understanding.

• Vision-Centric Prediction with a Standard Architecture (no task-specific modules): Youtu-VL treats image and text tokens with equivalent autoregressive status, empowering it to perform vision-centric tasks for both dense vision prediction (e.g., segmentation, depth) and text-based prediction (e.g., grounding, detection) within a standard VLM architecture, eliminating the need for task-specific additions. This design yields a versitile general-purpose VLM, allowing a single model to flexibly accommodate a wide range of vision-centric and vsion-language requirements.

🏆 Model Performance

Vision-Centric Tasks

General Multimodal Tasks

🚀 Quickstart

Using Transformers to Chat

Ensure your Python environment has the transformers library installed and that the version meets the requirements.

pip install "transformers>=4.56.0,<=4.57.1" torch accelerate pillow torchvision git+https://github.com/lucasb-eyer/pydensecrf.git opencv-python-headless

The snippet below shows how to interact with the chat model using transformers:

from transformers import AutoProcessor, AutoModelForCausalLM

model = AutoModelForCausalLM.from_pretrained(
    "tencent/Youtu-VL-4B-Instruct", attn_implementation="flash_attention_2", torch_dtype="auto", device_map="cuda", trust_remote_code=True
).eval()

processor = AutoProcessor.from_pretrained(
    "tencent/Youtu-VL-4B-Instruct", use_fast=True, trust_remote_code=True
)

img_path = "./assets/logo.png"
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "image": img_path},
            {"type": "text",  "text": "Describe the image"},
        ],
    }
]

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

generated_ids = model.generate(
    **inputs,
    temperature=0.1,
    top_p=0.001,
    repetition_penalty=1.05,
    do_sample=True,
    max_new_tokens=32768,
    img_input=img_path,
)

generated_ids_trimmed = [
    out_ids[len(in_ids) :] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)
]
outputs = processor.batch_decode(
    generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False
)
generated_text = outputs[0]
print(f"Youtu-VL output:\n{generated_text}")

Demo for VL and CV tasks

A simple demo for quick start, including VL and CV tasks.

cd demo
python demo.py

cd demo
jupyter notebook demo.ipynb

The core part of this demo is three lines below:

model_path = "tencent/Youtu-VL-4B-Instruct"
youtu_vl = YoutuVL(model_path)
response = youtu_vl(prompt, img_path, seg_mode=seg_mode)

Qualitative Results

• Task: Grounding

  Prompt: Please provide the bounding box coordinate of the region this sentence describes: a black and white cat sitting on the edge of the bathtub

  

• Task: Object Detection

  Prompt: Detect all objects in the provided image.

  

• Task: Referring Segmentation

  Prompt: Can you segment "hotdog on left" in this image?

  

For more examples, please refer to paper and Jupyter notebooks.

Using Llama.cpp to Chat

This guide will help you quickly deploy and invoke the Youtu-VL-4B-Instruct-GGUF model.

llama-server -hf tencent/Youtu-VL-4B-Instruct-GGUF:Q8_0  \
  --port 8080 \
  --image-max-tokens 2048 \
  --temp 0.1 \
  --top-p 0.001 \
  --repeat-penalty 1.05 \
  -n 12280 \
  --host 0.0.0.0

TODO List

• Release evaluation codes

🎉 Citation

If you find our work useful in your research, please consider citing our paper:

@article{youtu-vl,
  title={Youtu-VL: Unleashing Visual Potential via Unified Vision-Language Supervision},
  author={Tencent Youtu Lab},
  year={2026},
  eprint={2601.19798},
  archivePrefix={arXiv},
  primaryClass={cs.CV},
  url={https://arxiv.org/abs/2601.19798}, 
}

@article{youtu-llm,
  title={Youtu-LLM: Unlocking the Native Agentic Potential for Lightweight Large Language Models},
  author={Tencent Youtu Lab},
  year={2025},
  eprint={2512.24618},
  archivePrefix={arXiv},
  primaryClass={cs.CL},
  url={https://arxiv.org/abs/2512.24618}, 
}