RKNNAI/RK3576-VIT-clip-vit-base-patch32

clip-vit-base-patch32

English | 简体中文

1. Model Introduction

This repository provides RKNN deployment configurations of clip-vit-base-patch32 for RK3576.

Field	Value
Model ID	RKNNAI/RK3576-VIT-clip-vit-base-patch32
Display Name	RK3576-VIT-clip-vit-base-patch32
Source Model	https://huggingface.co/openai/clip-vit-base-patch32
Model Type	VIT

Available Models

Configuration	Supported Chips	RKNN Runtime Version	Quantization	NPU Cores	Resolution	Max Text Length
clip-vit-base-patch32-224x224-fp16-1	RK3576	v2.4.0	vision: fp16; text: fp16	vision: 1; text: 1	vision: 224x224	text: 20 (0.01953125k)

2. Files

Path	Description
README.md	English documentation
README_CN.md	Chinese documentation
LICENSE	License
clip-vit-base-patch32-224x224-fp16-1/	Deployment configuration

3. Download

ModelScope

Full repository:

modelscope download --model RKNNAI/RK3576-VIT-clip-vit-base-patch32 --revision v2.4.0 --local_dir ./RK3576-VIT-clip-vit-base-patch32

Selected configuration and root documents:

from modelscope import snapshot_download

snapshot_download(
    "RKNNAI/RK3576-VIT-clip-vit-base-patch32",
    revision="v2.4.0",
    allow_patterns=["README.md", "README_CN.md", "LICENSE", "clip-vit-base-patch32-224x224-fp16-1/**"],
    local_dir="./RK3576-VIT-clip-vit-base-patch32",
)

Hugging Face

Full repository:

hf download RKNNAI/RK3576-VIT-clip-vit-base-patch32 --revision v2.4.0 --local-dir ./RK3576-VIT-clip-vit-base-patch32

Selected configuration and root documents:

hf download RKNNAI/RK3576-VIT-clip-vit-base-patch32 --revision v2.4.0 --local-dir ./RK3576-VIT-clip-vit-base-patch32 --include README.md README_CN.md LICENSE 'clip-vit-base-patch32-224x224-fp16-1/**'

4. SHA-256 Verification

From the directory where you ran the download command, enter the configuration directory and verify:

cd ./RK3576-VIT-clip-vit-base-patch32/clip-vit-base-patch32-224x224-fp16-1
sha256sum -c SHA256SUMS

All entries must report OK before deployment.

5. Compatibility and Limitations

• Supported chips are listed per configuration above.
• Use the matching files from the same configuration.

6. Copyright and License

Source project: clip-vit-base-patch32. License: mit; see LICENSE. This distribution contains RKNN conversion derivatives for RK3576.