Latest AI Insights

Model Evaluations

Model Directory

Model Comparison

Resource Center

Search blog

加载中...

DataLearner AI

A knowledge platform focused on LLM benchmarking, datasets, and practical instruction with continuously updated capability maps.

产品

Leaderboards
模型对比
Datasets

资源

Tutorials
Editorial
Tool directory

关于

关于我们
隐私政策
数据收集方法
联系我们

© 2026 DataLearner AI. DataLearner curates industry data and case studies so researchers, enterprises, and developers can rely on trustworthy intelligence.

隐私政策服务条款

ViT-22B - AI Model Details and Benchmarks | DataLearnerAI

Page navigation

目录

Model catalogViT-22B

VI

ViT-22B

Vision Transformers-22B

Release date: 2023-02-10更新于: 2023-03-13 11:21:30.267566

Live demoGitHubHugging FaceCompare

Parameters

220.0亿

Context length

2K

Chinese support

Not supported

Reasoning ability

Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology

ViT-22B

Model basics

Reasoning traces

Not supported

Context length

2K tokens

Max output length

No data

Model type

基础大模型

Release date

2023-02-10

Model file size

ViT-22B

Open source & experience

Code license

No data

Weights license

No data

GitHub repo

GitHub link unavailable

Hugging Face

Hugging Face link unavailable

Live demo

No live demo

ViT-22B

Official resources

Paper

Scaling Vision Transformers to 22 Billion Parameters

DataLearnerAI blog

No blog post yet

ViT-22B

API details

API speed

No data

No public API pricing yet.

ViT-22B

Benchmark Results

No benchmark data to show.

ViT-22B

Publisher

Google Research

View publisher details

Vision Transformers-22B

Model Overview

ViT-22B是谷歌最新发布的220亿参数的视觉大模型。此前，虽然语言模型的参数一直增长，但是在视觉领域，模型的参数只有40亿。而往上的规模训练会出现一些不稳定情况。因此谷歌精心设计了ViT模型，使其有很高的并行训练效率。

DataLearner 官方微信

欢迎关注 DataLearner 官方微信，获得最新 AI 技术推送

DataLearner 官方微信二维码

No data

MoE architecture

No

Total params / Active params

220.0B / N/A

Knowledge cutoff

No data

Inference modes

No mode data