# xlm-roberta-large-xnli by Joe Davison: Open-Weight Model Source: https://savrn.com/models/xlm-roberta-large-xnli Markdown alternate of the page above; the site index is https://savrn.com/llms.txt --- ## Runs On What it takes to serve xlm-roberta-large-xnli (561M 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 | 1.1 GB | 1.3 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 8-bit | 0.6 GB | 0.7 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 4-bit | 0.3 GB | 0.3 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/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](https://savrn.com/ai-index/pricing/gpus), read Oct 22, 2026. [xlm-roberta-large-xnli on every accelerator the SAVRN Index prices, at every precision](https://savrn.com/models/xlm-roberta-large-xnli/gpus) ## SAVRN's Notes on xlm-roberta-large-xnli This is XLM-RoBERTa large fine-tuned on NLI data in 15 languages, and it was the model that made multilingual zero-shot classification practical in the Hugging Face pipeline. It has 561M parameters. The MIT license allows commercial use. Newer multilingual zero-shot models, such as Moritz Laurer's mDeBERTa and BGE-M3 versions, are usually more accurate. Keep this one where an existing system was tuned on it. SAVRN Research, 2026-09-18 ## Model Card By Joe Davison, published under mit, revision b227ee8435ce. ### Model Description This model takes [xlm-roberta-large](https://huggingface.co/xlm-roberta-large) and fine-tunes it on a combination of NLI data in 15 languages. It is intended to be used for zero-shot text classification, such as with the Hugging Face [ZeroShotClassificationPipeline](https://huggingface.co/transformers/master/main_classes/pipelines.html#transformers.ZeroShotClassificationPipeline). ### Intended Usage This model is intended to be used for zero-shot text classification, especially in languages other than English. It is fine-tuned on XNLI, which is a multilingual NLI dataset. The model can therefore be used with any of the languages in the XNLI corpus: - English - French - Spanish - German - Greek - Bulgarian - Russian - Turkish - Arabic - Vietnamese - Thai - Chinese - Hindi - Swahili - Urdu Since the base model was pre-trained trained on 100 different languages, the model has shown some effectiveness in languages beyond those listed above as well. See the full list of pre-trained languages in appendix A of the [XLM Roberata paper](https://savrn.com/papers/unsupervised-cross-lingual-representation-learning-at-scale) For English-only classification, it is recommended to use [bart-large-mnli](https://savrn.com/models/bart-large-mnli) or [a distilled bart MNLI model](https://huggingface.co/models?filter=pipeline_tag%3Azero-shot-classification&search=valhalla). #### With the zero-shot classification pipeline [Read the full model card (514 words)](https://savrn.com/models/xlm-roberta-large-xnli/card) ## Configuration Architecture XLMRobertaForSequenceClassification Context length (tokens) 514 Layers 24 Hidden size 1,024 Feed-forward size 4,096 Attention heads 16 Vocabulary size 250,002 Model type xlm-roberta ## Identity and Version Repository joeddav/xlm-roberta-large-xnli Publisher Joe Davison Task Zero-shot classification Modality Text Library transformers Parameters 561M parameters Languages en, fr, es, de, el, bg, ru, tr Revision b227ee8435ceadfa86dc1368a34254e2838bf242 First published 2022-03-02 Last updated 2024-10-16 ## Files and Weights 9 files, 6.7 GB in total. The weights are 3 files totalling 6.7 GB in bin, h5, safetensors. Weights3 files · 6.7 GB Configuration2 files · 884 B Tokenizer1 file · 25 B Documentation1 file · 5.2 KB Other1 file · 5.1 MB Repository1 file · 399 B Every file | File | Type | Size | SHA-256 | | --- | --- | --- | --- | | model.safetensors | Weights | 2.2 GB | 8869b0c99ad3 | | pytorch_model.bin | Weights | 2.2 GB | 947e0cebe098 | | tf_model.h5 | Weights | 2.2 GB | 913bd557db7f | | config.json | Configuration | 734 B | — | | special_tokens_map.json | Configuration | 150 B | — | | README.md | Documentation | 5.2 KB | — | | sentencepiece.bpe.model | Other | 5.1 MB | — | | .gitattributes | Repository | 399 B | — | | tokenizer_config.json | Tokenizer | 25 B | — | ## License and Download License mit Access Open weights, no gate Download size 6.7 GB [Download from Joe Davison](https://huggingface.co/joeddav/xlm-roberta-large-xnli) Released by Joe Davison through its official repository on Hugging Face. [Read the license](https://opensource.org/license/mit). ## Built From - Described by [arXiv:1911.02116](https://savrn.com/papers/unsupervised-cross-lingual-representation-learning-at-scale) - Trained on (disclosed) multi_nli - Trained on (disclosed) xnli ## Memory Requirements | Precision | Weights in memory | | --- | --- | | As published | 6.7 GB | | 16-bit | 1.1 GB | | 8-bit | 0.6 GB | | 4-bit | 0.3 GB | Weights only, from the published parameter count; the key-value cache and runtime add to this. ## Questions About xlm-roberta-large-xnli ### How much GPU memory does xlm-roberta-large-xnli need? About 1.3 GB at 16-bit and 0.3 GB at 4-bit: the weights (561M parameters) plus a working margin. A long context needs more. ### What is the cheapest GPU to run xlm-roberta-large-xnli 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. ### Can I use xlm-roberta-large-xnli commercially? Yes. xlm-roberta-large-xnli is released under MIT License. The MIT License is a short permissive license. It permits commercial use, modification and redistribution, provided the copyright notice and permission notice are included. ### What is xlm-roberta-large-xnli's context length? 514 tokens, from the maximum position embeddings in its published configuration. ## Similar Models Model · Zero-shot classification ### [bge-m3-zeroshot-v2.0](https://savrn.com/models/bge-m3-zeroshot-v2-0) [Moritz Borrett-Laurer (formerly Laurer)](https://savrn.com/model-publishers/moritzlaurer) Models in this series are designed for efficient zeroshot classification with the Hugging Face pipeline. These models can do classification without training data and run on both GPUs and CPUs. An overview of the latest zeroshot classifiers is available in my Zeroshot Classifier Collection. The main update of this zeroshot-v2.0 series of models is that several models are trained on fully commercially-friendly data for users with strict license requirements. These models can do one universal classification task: determine whether a hypothesis is "true" or "not true" given a text (entailment vs. notentailment). This task format is based on the Natural Language Inference task (NLI). The task is… Open weights mit 568M parameters 8,194 tokens transformers [View model](https://savrn.com/models/bge-m3-zeroshot-v2-0) Model · Zero-shot classification ### [LLM2CLIP-Openai-L-14-336](https://savrn.com/models/llm2clip-openai-l-14-336) [Microsoft](https://savrn.com/model-publishers/microsoft) Weiquan Huang 1, Aoqi Wu 1, Yifan Yang 2†, Xufang Luo 2, Yuqing Yang 2, Liang Hu 1, Qi Dai 2, Xiyang Dai 2, Dongdong Chen 2, Chong Luo 2, Lili Qiu 2 In this paper, we propose LLM2CLIP, a novel approach that embraces the power of LLMs to unlock CLIP’s potential. By fine-tuning the LLM in the caption space with contrastive learning, we extract its textual capabilities into the output embeddings, significantly improving the output layer’s textual discriminability. We then design an efficient training process where the fine-tuned LLM acts as a powerful teacher for CLIP’s visual encoder. Thanks to the LLM’s presence, we can now incorporate longer and more complex captions without being… Open weights apache-2.0 579M parameters [View model](https://savrn.com/models/llm2clip-openai-l-14-336) Model · Zero-shot classification ### [d3-edge](https://savrn.com/models/d3-edge) [vLLM Semantic Router](https://savrn.com/model-publishers/vllm-sr) d3-edge is the 0.6B multimodal foundation decision model of Decision 3.0, the decision models of vLLM Semantic Router. Give it an input (text or JSON, optionally with images) and the questions you need answered: pick one of several options, say yes or no, or rate on a scale. It answers them all in one call and returns a probability for every answer, without generating text. - Jev Decision Index 0.3, public suite: 30.55, measured with the official 0.3 kit on the released weights: all 140,178 public requests answered, none unsupported. - +13.9 on the public suite over Decision 2.0 (its 0.6B model: 16.68 on the board), ahead in all five areas. Images go before the text of the request, each… Open weights apache-2.0 659M parameters 262,144 tokens transformers [View model](https://savrn.com/models/d3-edge) Model · Zero-shot classification ### [Apex-flash](https://savrn.com/models/apex-flash) [SCX.ai Holdings Limited](https://savrn.com/model-publishers/scx-admin) Open weights apache-2.0 440M parameters gliclass [View model](https://savrn.com/models/apex-flash) Model · Zero-shot classification ### [DeBERTa-v3-large-mnli-fever-anli-ling-wanli](https://savrn.com/models/deberta-v3-large-mnli-fever-anli-ling-wanli) [Moritz Borrett-Laurer (formerly Laurer)](https://savrn.com/model-publishers/moritzlaurer) This model was fine-tuned on the MultiNLI, Fever-NLI, Adversarial-NLI (ANLI), LingNLI and WANLI datasets, which comprise 885 242 NLI hypothesis-premise pairs. This model is the best performing NLI model on the Hugging Face Hub as of 06.06.22 and can be used for zero-shot classification. It significantly outperforms all other large models on the ANLI benchmark. The foundation model is DeBERTa-v3-large from Microsoft. DeBERTa-v3 combines several recent innovations compared to classical Masked Language Models like BERT, RoBERTa etc., see the paper DeBERTa-v3-large-mnli-fever-anli-ling-wanli was trained on the MultiNLI, Fever-NLI, Adversarial-NLI (ANLI), LingNLI and WANLI datasets, which… Open weights mit 435M parameters 512 tokens transformers [View model](https://savrn.com/models/deberta-v3-large-mnli-fever-anli-ling-wanli) Model · Zero-shot classification ### [nli-deberta-v3-large](https://savrn.com/models/nli-deberta-v3-large) [Sentence Transformers - Cross-Encoders](https://savrn.com/model-publishers/cross-encoder) This model was trained using SentenceTransformers Cross-Encoder class. This model is based on microsoft/deberta-v3-large The model was trained on the SNLI and MultiNLI datasets. For a given sentence pair, it will output three scores corresponding to the labels: contradiction, entailment, neutral. For futher evaluation results, see SBERT.net - Pretrained Cross-Encoder. Pre-trained models can be used like this: You can use the model also directly with Transformers library (without SentenceTransformers library): This model can also be used for zero-shot-classification Open weights apache-2.0 435M parameters 512 tokens sentence-transformers [View model](https://savrn.com/models/nli-deberta-v3-large) ## Joe Davison [All models and datasets](https://savrn.com/model-publishers/joeddav) ## Versions - [b227ee8435ce](https://savrn.com/models/xlm-roberta-large-xnli/versions/b227ee8435ce) · current 2026-10-20 ## Explore More - [All zero-shot classification models](https://savrn.com/models/tasks/zero-shot-classification) - [All models under mit](https://savrn.com/models/licenses/mit) - [Model comparisons](https://savrn.com/models/comparisons) - [The model directory](https://savrn.com/models) - [Open model prices by host](https://savrn.com/ai-index/pricing/open-models) ## Source - Repository metadata, read 2026-10-19. - [Hugging Face record](https://huggingface.co/joeddav/xlm-roberta-large-xnli) - [How the hub is built](https://savrn.com/model-hub/methodology)