brightmart/roberta_zh resource

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

2791 stars · 408 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

A repository of pre-trained RoBERTa models for Chinese, implemented in TensorFlow with PyTorch versions also provided, loadable directly as BERT models. It includes multiple model sizes trained on ~30GB of Chinese text covering news, community Q&A, and encyclopedic data.

Use cases

• fine-tune a Chinese language model for text classification
• download a pretrained RoBERTa checkpoint for Chinese NLP tasks
• improve Chinese question answering with a stronger BERT-like encoder
• compare Chinese pretrained models like BERT, XLNet, and RoBERTa
• use a Chinese transformer model in TensorFlow or PyTorch

When to choose

• you need a strong pretrained Chinese encoder compatible with BERT loading code
• you work in TensorFlow or PyTorch and want ready-made Chinese model weights
• you want a smaller 6-layer or 12-layer model for faster experimentation

When to avoid

• you need a maintained library with active development and a license
• you need multilingual or English-only models
• you want the latest architectures like LLMs or instruction-tuned models

Facets

dataset · maturity maintenance

machine-learning nlp deep-learning machine-learning large-language-models python roberta chinese pretrained-models bert tensorflow pytorch model-weights natural-language-processing gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="brightmart/roberta_zh")