---
title: VisionHOPE
canonical_url: "https://www.modelscope.cn/models/PSRben/VisionHOPE"
md_url: "https://www.modelscope.cn/models/PSRben/VisionHOPE.md"
repository: PSRben/VisionHOPE
last_updated: 2026-09-30
license: mit
pipeline_tag: image-classification
tasks:
  - image-classification
library_name:
  - pytorch
frameworks:
  - pytorch
downloads: 0
stars: 0
tags:
  - computer-vision
  - pytorch
  - visionhope
---

# VisionHOPE

> VisionHOPE - PSRben 在 ModelScope 开源的模型。Official pretrained weights for VisionHOPE: Visual Backbones as Self-Modifying Learning Systems.

PSRben/VisionHOPE 是 ModelScope 魔搭社区上的image-classification模型，采用 mit 许可。

- **Repository**: PSRben/VisionHOPE
- **License**: mit
- **Tasks**: image-classification
- **Tags**: computer-vision, pytorch, visionhope
- **Downloads**: 0
- **Stars**: 0
- **Last updated**: 2026-09-30

Source: https://www.modelscope.cn/models/PSRben/VisionHOPE

---

# VisionHOPE

Official pretrained weights for **VisionHOPE: Visual Backbones as Self-Modifying Learning Systems**.

[Paper](https://arxiv.org/abs/2609.33325) · [Code](https://github.com/PSRben/VisionHOPE)

This repository contains the hierarchical VisionHOPE-T/S/B checkpoints for ImageNet-1K classification, COCO object detection and instance segmentation, and ADE20K semantic segmentation.

## Pretrained weights

| Task | VisionHOPE-T | VisionHOPE-S | VisionHOPE-B |
| :--- | :---: | :---: | :---: |
| ImageNet-1K classification | [Download](https://huggingface.co/PSRben/VisionHOPE/resolve/main/visionhope_tiny.pth?download=true) | [Download](https://huggingface.co/PSRben/VisionHOPE/resolve/main/visionhope_small.pth?download=true) | [Download](https://huggingface.co/PSRben/VisionHOPE/resolve/main/visionhope_base.pth?download=true) |
| COCO · Mask R-CNN 1× | [Download](https://huggingface.co/PSRben/VisionHOPE/resolve/main/visionhope_tiny_coco_1x.pth?download=true) | [Download](https://huggingface.co/PSRben/VisionHOPE/resolve/main/visionhope_small_coco_1x.pth?download=true) | [Download](https://huggingface.co/PSRben/VisionHOPE/resolve/main/visionhope_base_coco_1x.pth?download=true) |
| COCO · Mask R-CNN 3× | [Download](https://huggingface.co/PSRben/VisionHOPE/resolve/main/visionhope_tiny_coco_3x.pth?download=true) | [Download](https://huggingface.co/PSRben/VisionHOPE/resolve/main/visionhope_small_coco_3x.pth?download=true) | — |
| ADE20K · UPerNet | [Download](https://huggingface.co/PSRben/VisionHOPE/resolve/main/visionhope_tiny_ade20k.pth?download=true) | [Download](https://huggingface.co/PSRben/VisionHOPE/resolve/main/visionhope_small_ade20k.pth?download=true) | [Download](https://huggingface.co/PSRben/VisionHOPE/resolve/main/visionhope_base_ade20k.pth?download=true) |

## Download

Download an individual checkpoint with the [Hugging Face CLI](https://huggingface.co/docs/huggingface_hub/guides/cli):

```bash
hf download PSRben/VisionHOPE visionhope_small.pth --local-dir weights
```

To download all checkpoints:

```bash
hf download PSRben/VisionHOPE --include "visionhope_*.pth" --local-dir weights
```

See the [code repository](https://github.com/PSRben/VisionHOPE) for installation, model definitions, and training and evaluation commands.