source languages: fr; target languages: en; OPUS readme: fr-en; dataset: opus; model: transformer-align; pre-processing: normalization + SentencePiece.
transformer-en-zh is an open-weight model for translation from Chou Lucas. It has 98M parameters and a 5,000-token context. At 16-bit it needs about 0.2 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index.
基于标准 Transformer（Post-LN + 固定 sin-cos 位置编码）的英译中模型， 由本工程用 HuggingFace Seq2SeqTrainer 训练得到（checkpoint：trained with Seq2SeqTrainer (standard HF)）。 本仓库是标准 HF 仓库：包含 config.json + automap，代码随仓库分发， 因此任何人在 HF 上 trustremotecode=True 即可直接加载使用，无需本工程的其它文件。 逐步自回归推理，未实现 KV…
Parameters98M
Context5,000
Weights393.8 MB
License—
AccessOpen weights
Monthly Downloads—
Runs On
What it takes to serve transformer-en-zh (98M parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.
| Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits |
|---|---|---|---|---|---|
| 16-bit | 0.2 GB | 0.2 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 0.1 GB | 0.1 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 0.0 GB | 0.1 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Oct 18, 2026.
transformer-en-zh on every accelerator the SAVRN Index prices, at every precision
Model Card
基于标准 Transformer（Post-LN + 固定 sin-cos 位置编码）的英译中模型， 由本工程用 HuggingFace Seq2SeqTrainer 训练得到（checkpoint：trained with Seq2SeqTrainer (standard HF)）。 本仓库是标准 HF 仓库：包含 config.json + automap，代码随仓库分发， 因此任何人在 HF 上 trustremotecode=True 即可直接加载使用，无需本工程的其它文件。 逐步自回归推理，未实现 KV Cache（usecache=false），长句偏慢。 生成质量以 beam search（numbeams=3）优于贪心。 请按自身场景补充 LICENSE。
Excerpt from the card by Chou Lucas.
Configuration
- Architecture
- TransformerForConditionalGeneration
- Context length (tokens)
- 5,000
- Layers
- 6
- Vocabulary size
- 32,000
- Model type
- transformer_custom
Identity and Version
- Repository
- chou-lucas/transformer-en-zh
- Publisher
- Chou Lucas
- Task
- Translation
- Modality
- Text
- Library
- transformers
- Parameters
- 98M parameters
- Languages
- en, zh
- Revision
- 7d234b25380ce7995f95416b8b860448afc05001
- First published
- 2026-09-28
- Last updated
- 2026-09-28
Files and Weights
13 files, 397.1 MB in total. The weights are 1 file totalling 393.8 MB in safetensors.
Weights1 file · 393.8 MB
Configuration6 files · 955.3 KB
Tokenizer2 files · 750.7 KB
Documentation1 file · 2.6 KB
Other2 files · 1.6 MB
Repository1 file · 1.6 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model.safetensors | Weights | 393.8 MB | 248741cabfeb |
| config.json | Configuration | 1.0 KB | — |
| configuration_transformer.py | Configuration | 4.1 KB | — |
| generation_config.json | Configuration | 993 B | — |
| modeling_transformer.py | Configuration | 21.6 KB | — |
| target_vocab.json | Configuration | 920.5 KB | — |
| tokenization_transformer.py | Configuration | 7.2 KB | — |
| README.md | Documentation | 2.6 KB | — |
| source.spm | Other | 800.3 KB | 579e884a78d3 |
| target.spm | Other | 788.8 KB | 64f82f91e04c |
| .gitattributes | Repository | 1.6 KB | — |
| tokenizer_config.json | Tokenizer | 1.1 KB | — |
| vocab.json | Tokenizer | 749.6 KB | — |
License and Download
- License
- Not stated by the source
- Access
- Open weights, no gate
- Download size
- 393.8 MB
Download from Chou Lucas
Released by Chou Lucas through its official repository on Hugging Face.
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 393.8 MB |
| 16-bit | 0.2 GB |
| 8-bit | 0.1 GB |
| 4-bit | 0.0 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About transformer-en-zh
How much GPU memory does transformer-en-zh need?
About 0.2 GB at 16-bit and 0.1 GB at 4-bit: the weights (98M parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run transformer-en-zh on?
At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
What is transformer-en-zh's context length?
5,000 tokens, from the maximum position embeddings in its published configuration.
Similar Models
The developers of the Text-To-Text Transfer Transformer (T5) write: T5-Small is the checkpoint with 60 million parameters. The developers write in a blog post that the model: See the blog post and research paper for further details. The model is pre-trained on the Colossal Clean Crawled Corpus (C4), which was developed and released in the context of the same research paper as T5. The model was pre-trained on a on a multi-task mixture of unsupervised (1.) and supervised tasks (2.). Thereby, the following datasets were being used for (1.) and (2.): 1. Datasets used for Unsupervised denoising objective: 2. Datasets used for Supervised text-to-text language modeling objective - CoLA Warstadt et…
source languages: nl; target languages: en; OPUS readme: nl-en; dataset: opus; model: transformer-align; pre-processing: normalization + SentencePiece.
This is the model card of NLLB-200's distilled 600M variant. Here are the metrics for that particular checkpoint. - Information about training algorithms, parameters, fairness constraints or other applied approaches, and features. The exact training algorithm, data and the strategies to handle data imbalances for high and low resource languages that were used to train NLLB-200 is described in the paper. - Paper or other resource for more information NLLB Team et al, No Language Left Behind: Scaling Human-Centered Machine Translation, Arxiv, 2022 - Where to send questions or comments about the model: https://github.com/facebookresearch/fairseq/issues • Model performance measures: NLLB-200…
source languages: en; target languages: ru; OPUS readme: en-ru; dataset: opus; model: transformer-align; pre-processing: normalization + SentencePiece.
This model can be used for translation and text-to-text generation. CONTENT WARNING: Readers should be aware this section contains content that is disturbing, offensive, and can propagate historical and current stereotypes. Significant research has explored bias and fairness issues with language models (see, e.g., Sheng et al. (2021) and Bender et al. (2021)). Further details about the dataset for this model can be found in the OPUS readme: en-de