---
title: mobilebert-uncased
canonical_url: "https://www.modelscope.cn/models/google/mobilebert-uncased"
md_url: "https://www.modelscope.cn/models/google/mobilebert-uncased.md"
repository: google/mobilebert-uncased
last_updated: 2025-04-22
license: apache-2.0
model_type:
  - mobilebert
architectures:
  - MobileBertForPreTraining
library_name:
  - transformer
  - tensorflow
  - pytorch
frameworks:
  - pytorch
language:
  - en
downloads: 464
stars: 1
---

# mobilebert-uncased

> mobilebert-uncased - google 在 ModelScope 开源的模型。MobileBERT: a Compact Task-Agnostic BERT for Resource-Limited Devices

google/mobilebert-uncased 是 ModelScope 魔搭社区上的机器学习模型，采用 apache-2.0 许可。

- **Repository**: google/mobilebert-uncased
- **License**: apache-2.0
- **Downloads**: 464
- **Stars**: 1
- **Last updated**: 2025-04-22

Source: https://www.modelscope.cn/models/google/mobilebert-uncased

---

## MobileBERT: a Compact Task-Agnostic BERT for Resource-Limited Devices

MobileBERT is a thin version of BERT_LARGE, while equipped with bottleneck structures and a carefully designed balance
between self-attentions and feed-forward networks.

This checkpoint is the original MobileBert Optimized Uncased English: 
[uncased_L-24_H-128_B-512_A-4_F-4_OPT](https://storage.googleapis.com/cloud-tpu-checkpoints/mobilebert/uncased_L-24_H-128_B-512_A-4_F-4_OPT.tar.gz) 
checkpoint.

## How to use MobileBERT in `transformers`

```python
from transformers import pipeline

fill_mask = pipeline(
	"fill-mask",
	model="google/mobilebert-uncased",
	tokenizer="google/mobilebert-uncased"
)

print(
	fill_mask(f"HuggingFace is creating a {fill_mask.tokenizer.mask_token} that the community uses to solve NLP tasks.")
)

```