Bert Tiny Chinese is a language model designed for processing Traditional Chinese, made available in the ONNX format for use with ONNX Runtime. The model is based on the BERT architecture, specifically BertForMaskedLM, and it originates from the ckiplab/bert-tiny-chinese model. It has been converted to ONNX format to facilitate deployment and inference in environments that support ONNX.
The model features a compact architecture with 4 layers, a hidden size of 312, 12 attention heads, and an intermediate size of 1248. It supports a vocabulary size of 21,128 and can handle input sequences up to 512 tokens in length. The activation function used is GELU. export with dynamic axes for batch and sequence length.
0 license, allowing for open use and modification. The files provided include the full model in ONNX format (encoder and pooler) and a configuration file. This tool is suitable for developers or researchers who require a BERT-based language model for Traditional Chinese that can be easily integrated into ONNX-compatible workflows.
Bert Tiny Chinese is an Other AI product. It focuses on providing efficient and compact Chinese language understanding for NLP tasks using ONNX runtime. It is built as an open-source project for NLP developers and researchers working with Chinese text. Bert Tiny Chinese is open source under the GPL-3.0 license. Bert Tiny Chinese is available on the web and the command line, and it can be self-hosted.
Hosstia builds and maintains Bert Tiny Chinese, and the product first shipped in 2020. Development happens publicly on GitHub with 768 stars. PulseGate's similarity index finds few close equivalents — Bert Tiny Chinese occupies a relatively distinct niche. Key capabilities include chinese language support, ONNX format, and efficient inference.
Latest indexed changes and source events
Hosstia/bert-tiny-chinese-onnx discovered by the PulseGate indexer
Other apps tracked under the same category.
