This model is a conversion of MoritzLaurer/roberta-base-zeroshot-v2.0-c to ONNX format using the Optimum library.
Open-weight model · Zero-shot classification
tde-es-en-lite-plus
by Julián Araya jul879n/tde-es-en-lite-plus
tde-es-en-lite-plus is an open-weight model for zero-shot classification from Julián Araya, released under Creative Commons Attribution-NonCommercial 4.0. Its published files total 54.1 MB.
Variante plus (CC BY-NC 4.0, solo uso no comercial): ajustada con 15 datasets, varios de licencia no permisiva o no declarada. Para uso comercial existe tde-es-en-lite-v2 (MIT).
Parameters—
Context—
Weights53.0 MB
Licensecc-by-nc-4.0
AccessOpen weights
Monthly Downloads—
Model Card
Variante plus (CC BY-NC 4.0, solo uso no comercial): ajustada con 15 datasets, varios de licencia no permisiva o no declarada. Para uso comercial existe tde-es-en-lite-v2 (MIT). Motor de decisiones tipadas (choice / score / noul) sobre un encoder bidireccional int8 de ~42,5 M de parámetros (43 MB), recortado a español e inglés. No genera texto: elige entre opciones que le das, puntúa una rúbrica o estima si una proposición se desprende de un texto, y devuelve probabilidades, con la opción de abstenerse. Corre en CPU, 1 hilo, sin GPU y sin Python en runtime, con el binario nativo tde (Rust). Código fuente (Apache-2.0) incluido en src/; repositorio de desarrollo: typed-decision-engine.…
Excerpt from the card by Julián Araya, licensed cc-by-nc-4.0.
Identity and Version
- Repository
- jul879n/tde-es-en-lite-plus
- Publisher
- Julián Araya
- Task
- Zero-shot classification
- Modality
- Text
- Library
- onnxruntime
- Parameters
- Not stated by the source
- Languages
- es, en
- Revision
- f94be140cf4417ea0ad9574bbc7bf315423c54b5
- First published
- 2026-10-03
- Last updated
- 2026-10-03
Files and Weights
14 files, 54.1 MB in total. The weights are 4 files totalling 53.0 MB in bin, onnx.
Weights4 files · 53.0 MB
Configuration2 files · 4.0 KB
Documentation2 files · 11.6 KB
Other5 files · 1.1 MB
Repository1 file · 1.6 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| encoder_int8.onnx | Weights | 43.0 MB | 5bbde88eee8e |
| heads.bin | Weights | 7.1 MB | 9990dcf2f8b6 |
| tokenizer.bin | Weights | 1.3 MB | 52cb10d8ac99 |
| universal.bin | Weights | 1.6 MB | 6762135bb0ef |
| manifest.json | Configuration | 310 B | — |
| schemas/support.json | Configuration | 3.7 KB | — |
| LICENSE | Documentation | 4.3 KB | — |
| README.md | Documentation | 7.3 KB | — |
| bin/tde-macos-x86_64 | Other | 993.0 KB | 84d8663bf741 |
| install.sh | Other | 960 B | — |
| src/typed-decision-engine-src.tar.gz | Other | 58.6 KB | 2414d331c887 |
| tde | Other | 251 B | — |
| tools/fetch_ort.sh | Other | 3.3 KB | — |
| .gitattributes | Repository | 1.6 KB | — |
License and Download
- License
- cc-by-nc-4.0
- Access
- Open weights, no gate
- Download size
- 53.0 MB
Download from Julián Araya
Released by Julián Araya through its official repository on Hugging Face. Read the license.
Built From
- Derived from intfloat/multilingual-e5-small
- Quantized from intfloat/multilingual-e5-small
- Trained on (disclosed) DeepPavlov/hwu64
- Trained on (disclosed) SetFit/20_newsgroups
- Trained on (disclosed) SetFit/TREC-QC
- Trained on (disclosed) SetFit/bbc-news
- Trained on (disclosed) SetFit/sst5
- Trained on (disclosed) cardiffnlp/tweet_eval
- Trained on (disclosed) cardiffnlp/tweet_topic_single
- Trained on (disclosed) community-datasets/yahoo_answers_topics
- Trained on (disclosed) dair-ai/emotion
- Trained on (disclosed) fancyzhx/amazon_polarity
- Trained on (disclosed) fancyzhx/dbpedia_14
- Trained on (disclosed) fancyzhx/yelp_polarity
- Trained on (disclosed) heegyu/news-category-dataset
- Trained on (disclosed) mteb/banking77
- Trained on (disclosed) nyu-mll/multi_nli
- Trained on (disclosed) stanfordnlp/sst2
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 53.0 MB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About tde-es-en-lite-plus
Can I use tde-es-en-lite-plus commercially?
Not without separate permission. tde-es-en-lite-plus is released under Creative Commons Attribution-NonCommercial 4.0. CC BY-NC 4.0 permits sharing and adapting with credit for non-commercial purposes only. Commercial use needs separate permission from the rights holder.
Similar Models
https://huggingface.co/facebook/bart-large-mnli with ONNX weights to be compatible with Transformers.js. If you haven't already, you can install the Transformers.js JavaScript library from NPM using: Note: Having a separate repo for ONNX weights is intended to be a temporary solution until WebML gains more traction. If you would like to make your models web-ready, we recommend converting to ONNX using Optimum and structuring your repo like this one (with ONNX weights located in a subfolder named onnx).
distilbart-mnli is the distilled version of bart-large-mnli created using the No Teacher Distillation technique proposed for BART summarisation by Huggingface, here. We just copy alternating layers from bart-large-mnli and finetune more on the same data. This is a very simple and effective technique, as we can see the performance drop is very little. Detailed performace trade-offs will be posted in this sheet. If you want to train these models yourself, clone the distillbart-mnli repo and follow the steps below Clone and install transformers from source Download MNLI data Create student model Start fine-tuning You can find the logs of these trained models in this wandb project.
distilbart-mnli is the distilled version of bart-large-mnli created using the No Teacher Distillation technique proposed for BART summarisation by Huggingface, here. We just copy alternating layers from bart-large-mnli and finetune more on the same data. This is a very simple and effective technique, as we can see the performance drop is very little. Detailed performace trade-offs will be posted in this sheet. If you want to train these models yourself, clone the distillbart-mnli repo and follow the steps below Clone and install transformers from source Download MNLI data Create student model Start fine-tuning You can find the logs of these trained models in this wandb project.
https://huggingface.co/typeform/mobilebert-uncased-mnli with ONNX weights to be compatible with Transformers.js. If you haven't already, you can install the Transformers.js JavaScript library from NPM using: Note: Having a separate repo for ONNX weights is intended to be a temporary solution until WebML gains more traction. If you would like to make your models web-ready, we recommend converting to ONNX using Optimum and structuring your repo like this one (with ONNX weights located in a subfolder named onnx).
https://huggingface.co/cross-encoder/nli-deberta-v3-xsmall with ONNX weights to be compatible with Transformers.js. If you haven't already, you can install the Transformers.js JavaScript library from NPM using: Note: Having a separate repo for ONNX weights is intended to be a temporary solution until WebML gains more traction. If you would like to make your models web-ready, we recommend converting to ONNX using Optimum and structuring your repo like this one (with ONNX weights located in a subfolder named onnx).