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

ERNIE-4.5-VL-28B-A3B-Paddle

Model basics

Reasoning traces

Not supported

Context length

128K tokens

Max output length

32768 tokens

Model type

多模态大模型

Release date

2025-06-30

Model file size

58.84 GB

MoE architecture

Yes

Total params / Active params

280.0B / 30B

Knowledge cutoff

No data

Inference modes

No mode data

ERNIE-4.5-VL-28B-A3B-Paddle

Open source & experience

Code license

Apache 2.0

Weights license

Apache 2.0- 免费商用授权

GitHub repo

GitHub link unavailable

Hugging Face

https://huggingface.co/baidu/ERNIE-4.5-VL-28B-A3B-Paddle

Live demo

https://yiyan.baidu.com/

ERNIE-4.5-VL-28B-A3B-Paddle

Official resources

Paper

No paper available

DataLearnerAI blog

No blog post yet

ERNIE-4.5-VL-28B-A3B-Paddle

API details

API speed

3/5

No public API pricing yet.

ERNIE-4.5-VL-28B-A3B-Paddle

Benchmark Results

No benchmark data to show.

ERNIE-4.5-VL-28B-A3B-Paddle

Publisher

百度

View publisher details

ERNIE-4.5-VL-28B-A3B-Paddle

Model Overview

ERNIE-4.5-VL-28B-A3B-Paddle是百度开源的MoE架构多模态大模型，总参数280亿，激活参数30亿。

DataLearner 官方微信

欢迎关注 DataLearner 官方微信，获得最新 AI 技术推送