# DeBERTa-v3-base-mnli-fever-anli by Moritz Borrett-Laurer Source: https://savrn.com/models/deberta-v3-base-mnli-fever-anli Markdown alternate of the page above; the site index is https://savrn.com/llms.txt --- ## Runs On What it takes to serve DeBERTa-v3-base-mnli-fever-anli (184M 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 | 0.4 GB | 0.4 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 | 0.2 GB | 0.2 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 | 0.1 GB | 0.1 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 23, 2026. [DeBERTa-v3-base-mnli-fever-anli on every accelerator the SAVRN Index prices, at every precision](https://savrn.com/models/deberta-v3-base-mnli-fever-anli/gpus) ## SAVRN's Notes on DeBERTa-v3-base-mnli-fever-anli Give this 184M parameter model a premise and a hypothesis and it scores whether one follows from the other, which is the mechanism behind sorting unlabeled text into categories you name yourself. At 16-bit the weights take 0.4 GB, 8-bit 0.2 GB, 4-bit 0.1 GB. The cheapest card on our Index is one MI300X with 192 GB at $1.85 per hour on-demand, and this model would use a fraction of one percent of it; plan other work on the same card. MIT permits commercial use, modification and redistribution provided the copyright and permission notices travel with the files. The context window is 512 tokens, so long documents get chunked upstream. The 0.737 ANLI accuracy is the publisher's own number. The publisher also points to the DeBERTa-v3-large variant for highest performance at the cost of speed, so pick a side of that trade before sizing the deployment. SAVRN Research, 2026-09-18 ## Model Card By Moritz Borrett-Laurer (formerly Laurer), published under mit, revision 6f5cf0a2b59c. ### Model description This model was trained on the MultiNLI, Fever-NLI and Adversarial-NLI (ANLI) datasets, which comprise 763 913 NLI hypothesis-premise pairs. This base model outperforms almost all large models on the [ANLI benchmark](https://github.com/facebookresearch/anli). The base model is [DeBERTa-v3-base from Microsoft](https://savrn.com/models/deberta-v3-base). The v3 variant of DeBERTa substantially outperforms previous versions of the model by including a different pre-training objective, see annex 11 of the original [DeBERTa paper](https://savrn.com/papers/deberta-decoding-enhanced-bert-with-disentangled-attention). For highest performance (but less speed), I recommend using https://huggingface.co/MoritzLaurer/DeBERTa-v3-large-mnli-fever-anli-ling-wanli. #### How to use the model ##### Simple zero-shot classification pipeline ``` #!pip install transformers[sentencepiece] from transformers import pipeline classifier = pipeline("zero-shot-classification", model="MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli") sequence_to_classify = "Angela Merkel is a politician in Germany and leader of the CDU" candidate_labels = ["politics", "economy", "entertainment", "environment"] output = classifier(sequence_to_classify, candidate_labels, multi_label=False) print(output) ``` ##### NLI use-case [Read the full model card (586 words)](https://savrn.com/models/deberta-v3-base-mnli-fever-anli/card) ## Configuration Architecture DebertaV2ForSequenceClassification Context length (tokens) 512 Layers 12 Hidden size 768 Feed-forward size 3,072 Attention heads 12 Vocabulary size 128,100 Stored precision float16 Model type deberta-v2 ## Identity and Version Repository MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli Publisher Moritz Borrett-Laurer (formerly Laurer) Task Zero-shot classification Modality Text Library transformers Parameters 184M parameters Languages en Revision 6f5cf0a2b59cabb106aca4c287eed12e357e90eb First published 2022-03-02 Last updated 2024-04-11 ## Files and Weights 10 files, 749.0 MB in total. The weights are 2 files totalling 737.8 MB in bin, safetensors. Weights2 files · 737.8 MB Configuration3 files · 1.4 KB Tokenizer2 files · 8.7 MB Documentation1 file · 23.6 KB Other1 file · 2.5 MB Repository1 file · 1.2 KB Every file | File | Type | Size | SHA-256 | | --- | --- | --- | --- | | model.safetensors | Weights | 368.9 MB | 06d6fd89edd4 | | pytorch_model.bin | Weights | 368.9 MB | 2a2d9e8fc302 | | added_tokens.json | Configuration | 23 B | — | | config.json | Configuration | 1.1 KB | — | | special_tokens_map.json | Configuration | 286 B | — | | README.md | Documentation | 23.6 KB | — | | spm.model | Other | 2.5 MB | c679fbf93643 | | .gitattributes | Repository | 1.2 KB | — | | tokenizer.json | Tokenizer | 8.7 MB | — | | tokenizer_config.json | Tokenizer | 1.3 KB | — | ## License and Download License mit Access Open weights, no gate Download size 737.8 MB [Download from Moritz Borrett-Laurer (formerly Laurer)](https://huggingface.co/MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli) Released by Moritz Borrett-Laurer (formerly Laurer) through its official repository on Hugging Face. [Read the license](https://opensource.org/license/mit). ## Built From - Described by [arXiv:2006.03654](https://savrn.com/papers/deberta-decoding-enhanced-bert-with-disentangled-attention) - Trained on (disclosed) [facebook/anli](https://savrn.com/datasets/anli) - Trained on (disclosed) fever - Trained on (disclosed) multi_nli ## Evaluations Each result is shown as reported, with the conditions its reporter stated. None is a SAVRN measurement. A comparison lines two results up only when their configuration, unit and setup are all stated and identical. | Benchmark | Conditions | Result | Reported by | Revision | Date | | --- | --- | --- | --- | --- | --- | | anli | Configuration plain_textTask Natural Language InferenceMetric AccuracyComparison conditions not established | 0.737 | [MoritzLaurer](https://huggingface.co/MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli) Publisher reported | Evaluated revision not stated | — | | anli | Configuration plain_textTask Natural Language InferenceMetric F1 MacroComparison conditions not established | 0.711923 | [MoritzLaurer](https://huggingface.co/MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli) Publisher reported | Evaluated revision not stated | — | | anli | Configuration plain_textTask Natural Language InferenceMetric F1 MicroComparison conditions not established | 0.712 | [MoritzLaurer](https://huggingface.co/MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli) Publisher reported | Evaluated revision not stated | — | | anli | Configuration plain_textTask Natural Language InferenceMetric F1 WeightedComparison conditions not established | 0.711924 | [MoritzLaurer](https://huggingface.co/MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli) Publisher reported | Evaluated revision not stated | — | | anli | Configuration plain_textTask Natural Language InferenceMetric Precision MacroComparison conditions not established | 0.737681 | [MoritzLaurer](https://huggingface.co/MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli) Publisher reported | Evaluated revision not stated | — | | anli | Configuration plain_textTask Natural Language InferenceMetric Precision MicroComparison conditions not established | 0.737 | [MoritzLaurer](https://huggingface.co/MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli) Publisher reported | Evaluated revision not stated | — | | anli | Configuration plain_textTask Natural Language InferenceMetric Precision WeightedComparison conditions not established | 0.737676 | [MoritzLaurer](https://huggingface.co/MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli) Publisher reported | Evaluated revision not stated | — | | anli | Configuration plain_textTask Natural Language InferenceMetric Recall MacroComparison conditions not established | 0.736968 | [MoritzLaurer](https://huggingface.co/MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli) Publisher reported | Evaluated revision not stated | — | | anli | Configuration plain_textTask Natural Language InferenceMetric Recall MicroComparison conditions not established | 0.737 | [MoritzLaurer](https://huggingface.co/MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli) Publisher reported | Evaluated revision not stated | — | | anli | Configuration plain_textTask Natural Language InferenceMetric Recall WeightedComparison conditions not established | 0.737 | [MoritzLaurer](https://huggingface.co/MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli) Publisher reported | Evaluated revision not stated | — | | anli | Configuration plain_textTask Natural Language InferenceMetric lossComparison conditions not established | 1.01054 | [MoritzLaurer](https://huggingface.co/MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli) Publisher reported | Evaluated revision not stated | — | | multi_nli | Configuration defaultTask Natural Language InferenceMetric AccuracyComparison conditions not established | 0.902766 | [MoritzLaurer](https://huggingface.co/MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli) Publisher reported | Evaluated revision not stated | — | | multi_nli | Configuration defaultTask Natural Language InferenceMetric F1 MacroComparison conditions not established | 0.902309 | [MoritzLaurer](https://huggingface.co/MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli) Publisher reported | Evaluated revision not stated | — | | multi_nli | Configuration defaultTask Natural Language InferenceMetric F1 MicroComparison conditions not established | 0.902766 | [MoritzLaurer](https://huggingface.co/MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli) Publisher reported | Evaluated revision not stated | — | | multi_nli | Configuration defaultTask Natural Language InferenceMetric F1 WeightedComparison conditions not established | 0.903016 | [MoritzLaurer](https://huggingface.co/MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli) Publisher reported | Evaluated revision not stated | — | | multi_nli | Configuration defaultTask Natural Language InferenceMetric Precision MacroComparison conditions not established | 0.902382 | [MoritzLaurer](https://huggingface.co/MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli) Publisher reported | Evaluated revision not stated | — | | multi_nli | Configuration defaultTask Natural Language InferenceMetric Precision MicroComparison conditions not established | 0.902766 | [MoritzLaurer](https://huggingface.co/MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli) Publisher reported | Evaluated revision not stated | — | | multi_nli | Configuration defaultTask Natural Language InferenceMetric Precision WeightedComparison conditions not established | 0.90346 | [MoritzLaurer](https://huggingface.co/MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli) Publisher reported | Evaluated revision not stated | — | | multi_nli | Configuration defaultTask Natural Language InferenceMetric Recall MacroComparison conditions not established | 0.90243 | [MoritzLaurer](https://huggingface.co/MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli) Publisher reported | Evaluated revision not stated | — | | multi_nli | Configuration defaultTask Natural Language InferenceMetric Recall MicroComparison conditions not established | 0.902766 | [MoritzLaurer](https://huggingface.co/MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli) Publisher reported | Evaluated revision not stated | — | | multi_nli | Configuration defaultTask Natural Language InferenceMetric Recall WeightedComparison conditions not established | 0.902766 | [MoritzLaurer](https://huggingface.co/MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli) Publisher reported | Evaluated revision not stated | — | | multi_nli | Configuration defaultTask Natural Language InferenceMetric lossComparison conditions not established | 0.328335 | [MoritzLaurer](https://huggingface.co/MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli) Publisher reported | Evaluated revision not stated | — | ## Memory Requirements | Precision | Weights in memory | | --- | --- | | As published | 737.8 MB | | 16-bit | 0.4 GB | | 8-bit | 0.2 GB | | 4-bit | 0.1 GB | Weights only, from the published parameter count; the key-value cache and runtime add to this. ## Questions About DeBERTa-v3-base-mnli-fever-anli ### How much GPU memory does DeBERTa-v3-base-mnli-fever-anli need? About 0.4 GB at 16-bit and 0.1 GB at 4-bit: the weights (184M parameters) plus a working margin. A long context needs more. ### What is the cheapest GPU to run DeBERTa-v3-base-mnli-fever-anli 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 DeBERTa-v3-base-mnli-fever-anli commercially? Yes. DeBERTa-v3-base-mnli-fever-anli 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 DeBERTa-v3-base-mnli-fever-anli's context length? 512 tokens, from the maximum position embeddings in its published configuration. ## Similar Models Model · Zero-shot classification ### [nli-deberta-v3-base](https://savrn.com/models/nli-deberta-v3-base) [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-base 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 184M parameters 512 tokens sentence-transformers [View model](https://savrn.com/models/nli-deberta-v3-base) Model · Zero-shot classification ### [deberta-v3-base-tasksource-nli](https://savrn.com/models/deberta-v3-base-tasksource-nli) [Damien Sileo](https://savrn.com/model-publishers/sileod) NOTE Deprecated: use https://huggingface.co/tasksource/deberta-small-long-nli for longer context and better accuracy. This is DeBERTa-v3-base fine-tuned with multi-task learning on 600+ tasks of the tasksource collection. This checkpoint has strong zero-shot validation performance on many tasks (e.g. 70% on WNLI), and can be used for: - Zero-shot entailment-based classification for arbitrary labels [ZS]. - Natural language inference [NLI] - Hundreds of previous tasks with tasksource-adapters [TA]. - Further fine-tuning on a new task or tasksource task (classification, token classification or multiple-choice) [FT]. NLI training data of this model includes label-nli, a NLI dataset specially… Open weights apache-2.0 184M parameters 512 tokens transformers [View model](https://savrn.com/models/deberta-v3-base-tasksource-nli) Model · Zero-shot classification ### [DeBERTa-v3-base-mnli](https://savrn.com/models/deberta-v3-base-mnli) [Moritz Borrett-Laurer (formerly Laurer)](https://savrn.com/model-publishers/moritzlaurer) This model was trained on the MultiNLI dataset, which consists of 392 702 NLI hypothesis-premise pairs. The base model is DeBERTa-v3-base from Microsoft. The v3 variant of DeBERTa substantially outperforms previous versions of the model by including a different pre-training objective, see annex 11 of the original DeBERTa paper. For a more powerful model, check out DeBERTa-v3-base-mnli-fever-anli which was trained on even more data. This model was trained on the MultiNLI dataset, which consists of 392 702 NLI hypothesis-premise pairs. DeBERTa-v3-base-mnli was trained using the Hugging Face trainer with the following hyperparameters. The model was evaluated using the matched test set and… Open weights 184M parameters 512 tokens transformers [View model](https://savrn.com/models/deberta-v3-base-mnli) Model · Zero-shot classification ### [deberta-base-long-nli](https://savrn.com/models/deberta-base-long-nli) [Tasksource](https://savrn.com/model-publishers/tasksource) deberta-v3-base with context length of 1280 fine-tuned on tasksource for 250k steps. I oversampled long NLI tasks (ConTRoL, doc-nli). Training data include helpsteer v1/v2, logical reasoning tasks (FOLIO, FOL-nli, LogicNLI...), OASST, hh/rlhf, linguistics oriented NLI tasks, tasksource-dpo, fact verification tasks. This checkpoint has strong zero-shot validation performance on many tasks (e.g. 70% on WNLI), and can be used for: - Zero-shot entailment-based classification for arbitrary labels [ZS]. - Natural language inference [NLI] - Further fine-tuning on a new task or tasksource task (classification, token classification, reward modeling or multiple-choice) [FT]. Zero-shot GPT-4 scores… Open weights apache-2.0 184M parameters 1,280 tokens transformers [View model](https://savrn.com/models/deberta-base-long-nli) Model · Zero-shot classification ### [deberta-v3-base-zeroshot-v2.0](https://savrn.com/models/deberta-v3-base-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 184M parameters 512 tokens transformers [View model](https://savrn.com/models/deberta-v3-base-zeroshot-v2-0) Model · Zero-shot classification ### [deberta-v3-base-zeroshot-v1.1-all-33](https://savrn.com/models/deberta-v3-base-zeroshot-v1-1-all-33) [Moritz Borrett-Laurer (formerly Laurer)](https://savrn.com/model-publishers/moritzlaurer) The model is designed for zero-shot classification with the Hugging Face pipeline. The model 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 so universal that any classification task can be reformulated into this task. A detailed description of how the model was trained and how it can be used is available in this paper. The model was trained on a mixture of 33 datasets and 387 classes that have been reformatted into this universal format. 1. Five NLI datasets with ~885k texts: "mnli", "anli", "fever", "wanli"… Open weights mit 184M parameters 512 tokens transformers [View model](https://savrn.com/models/deberta-v3-base-zeroshot-v1-1-all-33) ## Moritz Borrett-Laurer (formerly Laurer) [All models and datasets](https://savrn.com/model-publishers/moritzlaurer) ## Versions - [6f5cf0a2b59c](https://savrn.com/models/deberta-v3-base-mnli-fever-anli/versions/6f5cf0a2b59c) · current 2026-09-18 ## 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 2026-10-20. - [Hugging Face record](https://huggingface.co/MoritzLaurer/DeBERTa-v3-base-mnli-fever-anli) - [How the hub is built](https://savrn.com/model-hub/methodology)