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
nli-deberta-v3-small
by Joshua Xenova/nli-deberta-v3-small
https://huggingface.co/cross-encoder/nli-deberta-v3-small with ONNX weights to be compatible with Transformers.js.
Parameters—
Context512
Weights2.5 GB
License—
AccessOpen weights
Monthly Downloads2.1k
Model Card
https://huggingface.co/cross-encoder/nli-deberta-v3-small 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).
Excerpt from the card by Joshua.
Configuration
- Architecture
- DebertaV2ForSequenceClassification
- Context length (tokens)
- 512
- Layers
- 6
- Hidden size
- 768
- Feed-forward size
- 3,072
- Attention heads
- 12
- Vocabulary size
- 128,100
- Model type
- deberta-v2
Identity and Version
- Repository
- Xenova/nli-deberta-v3-small
- Publisher
- Joshua
- Task
- Zero-shot classification
- Modality
- Text
- Library
- transformers.js
- Parameters
- Not stated by the source
- Languages
- Not stated by the source
- Revision
- 6bc2a55c7c0f7e2bc68de60bb248e523e2612abb
- First published
- 2023-08-08
- Last updated
- 2025-07-14
Files and Weights
17 files, 2.5 GB in total. The weights are 8 files totalling 2.5 GB in onnx.
Weights8 files · 2.5 GB
Configuration4 files · 2.4 KB
Tokenizer2 files · 8.7 MB
Documentation1 file · 1.2 KB
Other1 file · 2.5 MB
Repository1 file · 1.5 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| onnx/model.onnx | Weights | 568.0 MB | 59e082152f6e |
| onnx/model_bnb4.onnx | Weights | 439.2 MB | 9783afcb918a |
| onnx/model_fp16.onnx | Weights | 284.4 MB | d2962e282cde |
| onnx/model_int8.onnx | Weights | 160.9 MB | fda286f22872 |
| onnx/model_q4.onnx | Weights | 441.9 MB | 516132fc6e6f |
| onnx/model_q4f16.onnx | Weights | 231.7 MB | b067e9594400 |
| onnx/model_quantized.onnx | Weights | 172.4 MB | 4edae6adf6ef |
| onnx/model_uint8.onnx | Weights | 160.9 MB | aebbb0ecfb03 |
| added_tokens.json | Configuration | 23 B | — |
| config.json | Configuration | 1.0 KB | — |
| quantize_config.json | Configuration | 1.2 KB | — |
| special_tokens_map.json | Configuration | 173 B | — |
| README.md | Documentation | 1.2 KB | — |
| spm.model | Other | 2.5 MB | c679fbf93643 |
| .gitattributes | Repository | 1.5 KB | — |
| tokenizer.json | Tokenizer | 8.7 MB | — |
| tokenizer_config.json | Tokenizer | 384 B | — |
License and Download
- License
- Not stated by the source
- Access
- Open weights, no gate
- Download size
- 2.5 GB
Download from Joshua
Released by Joshua through its official repository on Hugging Face.
Built From
- Derived from cross-encoder/nli-deberta-v3-small
- Quantized from cross-encoder/nli-deberta-v3-small
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 2.5 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About nli-deberta-v3-small
What is nli-deberta-v3-small's context length?
512 tokens, from the maximum position embeddings in its published configuration.
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).