---
title: clip
canonical_url: "https://www.modelscope.cn/models/AXERA-TECH/clip"
md_url: "https://www.modelscope.cn/models/AXERA-TECH/clip.md"
repository: AXERA-TECH/clip
last_updated: 2026-07-18
license: mit
model_type:
  - internvl_chat
architectures:
  - InternVLChatModel
library_name:
  - onnx
  - pytorch
frameworks:
  - pytorch
downloads: 61
stars: 0
tags:
  - Transformer
  - ONNX
  - clip
---

# clip

> clip - AXERA-TECH 在 ModelScope 开源的模型。Original Repository: openai/CLIP

AXERA-TECH/clip 是 ModelScope 魔搭社区上的机器学习模型，采用 mit 许可。

- **Repository**: AXERA-TECH/clip
- **License**: mit
- **Tags**: Transformer, ONNX, clip
- **Downloads**: 61
- **Stars**: 0
- **Last updated**: 2026-07-18

Source: https://www.modelscope.cn/models/AXERA-TECH/clip

---

# CLIP

- [Original Repository: openai/CLIP](https://github.com/openai/CLIP)

## How to convert ONNX to axmodel 

For those who are interested in model conversion, you can try to export axmodel through the original repo: 

- [clip.axera: export onnx from clip repo](https://github.com/AXERA-TECH/clip.axera)

- [Pulsar2 Link: How to Convert from ONNX to axmodel](https://pulsar2-docs.readthedocs.io/en/latest/) 

## How to deploy the axmodel

- [AXera NPU HOST Runtime](https://github.com/AXERA-TECH/CLIP-ONNX-AX650-CPP) 

## Support Platform

- AX650
  - [M4N-Dock(爱芯派Pro)](https://wiki.sipeed.com/hardware/zh/maixIV/m4ndock/m4ndock.html)
  - [M.2 Accelerator card](https://axcl-docs.readthedocs.io/zh-cn/latest/doc_guide_hardware.html)

## Inference with AX650 Host

- check the [on board reference](https://github.com/AXERA-TECH/CLIP-ONNX-AX650-CPP) for more information