---
title: LoopVL
canonical_url: "https://www.modelscope.cn/models/Eternity123/LoopVL"
md_url: "https://www.modelscope.cn/models/Eternity123/LoopVL.md"
repository: Eternity123/LoopVL
last_updated: 2026-10-01
pipeline_tag: image-text-to-text
tasks:
  - image-text-to-text
model_type:
  - loopvl
architectures:
  - LoopVLForConditionalGeneration
parameters: 1.6B
tensor_type:
  - BF16
  - F32
library_name:
  - safetensors
downloads: 3
stars: 0
tags:
  - loopvl
  - vision-language
---

# LoopVL

> LoopVL - Eternity123 在 ModelScope 开源的模型。Vision-language modeling with recurrent visual computation GitHub · Code & evaluation  ·  Hugging Face  ·  ModelScope

- **Repository**: Eternity123/LoopVL
- **Tasks**: image-text-to-text
- **Parameters**: 1.6B
- **Tags**: loopvl, vision-language
- **Downloads**: 3
- **Stars**: 0
- **Last updated**: 2026-10-01

Source: https://www.modelscope.cn/models/Eternity123/LoopVL

---

<p align="center">
  <img src="https://raw.githubusercontent.com/Tier-Flow/LoopVL/main/assets/loopvl-wordmark.svg" width="520" alt="LoopVL">
</p>
<p align="center"><strong>Vision-language modeling with recurrent visual computation</strong></p>
<p align="center">
  <a href="https://github.com/Tier-Flow/LoopVL">GitHub · Code &amp; evaluation</a> &nbsp;·&nbsp;
  <a href="https://huggingface.co/TierFlow/LoopVL">Hugging Face</a> &nbsp;·&nbsp;
  <a href="https://modelscope.cn/models/Eternity123/LoopVL">ModelScope</a>
</p>

This repository contains **one `model.safetensors` and flat configuration,
image-processor and tokenizer files**. All inference and evaluation code is
kept in the [LoopVL GitHub repository](https://github.com/Tier-Flow/LoopVL).
There are no Python files in this model snapshot.

## Quick start

Use Linux, Python 3.12 and a matching CUDA-compatible PyTorch/torchvision pair.
The validated GPU environment uses torch 2.12.1 and torchvision 0.27.1.

```bash
git clone https://github.com/Tier-Flow/LoopVL.git
cd LoopVL
python -m pip install -r requirements.txt

hf download TierFlow/LoopVL --local-dir model
# Alternative: ms-hub download Eternity123/LoopVL --local-dir model

python scripts/verify_repo.py --verify-model
python runtime/infer.py --image /path/to/image.png --prompt "What is in the image?" --budget 32 --device cuda:0
```

Keep **both the GitHub code and all downloaded model files**. LoopVL uses its
custom GitHub loader.

The main checkpoint uses H2L3: `L → L → L → H → L → L → L → H`, with 16 layers
per module call and **128 effective layer applications per forward pass**.
Preserve its mixed BF16 core weights and FP32 adapters; do not cast the entire
model with `.half()` or `.bfloat16()`.

> [!TIP]
> For LoopVL-1B benchmark evaluation and behavior exploration, prefer direct
> answers. Chain-of-thought prompting is recommended only for mathematics tasks.