# SecBERT by Kun: Open-Weight Model Source: https://savrn.com/models/secbert Markdown alternate of the page above; the site index is https://savrn.com/llms.txt --- ## Runs On What it takes to serve SecBERT (84M 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](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 | 0.1 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 | | 4-bit | 0.0 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. [SecBERT on every accelerator the SAVRN Index prices, at every precision](https://savrn.com/models/secbert/gpus) ## SAVRN's Notes on SecBERT Security text has its own vocabulary, and this model was built around that fact: a 52,000-entry wordpiece vocabulary the publisher calls secvocab, matched to a corpus of cyber security papers. The fill-mask head is the starting point, not the product; the publisher positions it as a base for NER, text classification and Q&A. Six layers and 84M parameters come to 0.2 GB at 16-bit, so the cheapest Index setup, one MI300X with 192 GB at $1.85 per hour on-demand, is a card to share across many fine-tunes, not dedicate to one. Apache 2.0 means a fine-tuned derivative can ship commercially, redistributed, with the notices kept and changes stated. Check the 514-token context first: documents longer than that get chunked. Then compare your text against the training sets, APTnotes, Stucco-Data and CASIE, and note the publisher also trained a SecRoBERTa version and last updated this one on 2023-06-26. SAVRN Research, 2026-09-18 ## Model Card By Kun, published under apache-2.0, revision 7c603df5bc4c. This is the pretrained model presented in SecBERT: A Pretrained Language Model for Cyber Security Text, which is a BERT model trained on cyber security text. The training corpus was papers taken from SecBERT has its own wordpiece vocabulary (secvocab) that's built to best match the training corpus. We trained SecBERT and SecRoBERTa versions. We proposed to build language model which work on cyber security text, as result, it can improve downstream tasks (NER, Text Classification, Semantic Understand, Q&A) in Cyber Security Domain. First, as below shows Fill-Mask pipeline in Google Bert, AllenAI SciBert and our SecBERT. The original repo can be found here. Read Kun's full model card This is the pretrained model presented in [SecBERT: A Pretrained Language Model for Cyber Security Text](https://github.com/jackaduma/SecBERT/), which is a BERT model trained on cyber security text. The training corpus was papers taken from * [APTnotes](https://github.com/kbandla/APTnotes) * [Stucco-Data: Cyber security data sources](https://stucco.github.io/data/) * [CASIE: Extracting Cybersecurity Event Information from Text](https://ebiquity.umbc.edu/_file_directory_/papers/943.pdf) * [SemEval-2018 Task 8: Semantic Extraction from CybersecUrity REports using Natural Language Processing (SecureNLP)](https://competitions.codalab.org/competitions/17262). SecBERT has its own wordpiece vocabulary (secvocab) that's built to best match the training corpus. We trained [SecBERT](https://savrn.com/models/secbert) and [SecRoBERTa](https://huggingface.co/jackaduma/SecRoBERTa) versions. Available models include: * [SecBERT](https://savrn.com/models/secbert) * [SecRoBERTa](https://huggingface.co/jackaduma/SecRoBERTa) ### Fill Mask We proposed to build language model which work on cyber security text, as result, it can improve downstream tasks (NER, Text Classification, Semantic Understand, Q&A) in Cyber Security Domain. First, as below shows Fill-Mask pipeline in Google Bert, [AllenAI SciBert](https://github.com/allenai/scibert) and our [SecBERT](https://github.com/jackaduma/SecBERT) . The original repo can be found [here](https://github.com/jackaduma/SecBERT). ## Configuration Architecture BertForMaskedLM Context length (tokens) 514 Layers 6 Hidden size 768 Feed-forward size 3,072 Attention heads 12 Vocabulary size 52,000 Model type bert ## Identity and Version Repository jackaduma/SecBERT Publisher Kun Task Fill mask Modality Text Library transformers Parameters 84M parameters Languages en Revision 7c603df5bc4c5ba9c731bcc2ea0ab2db36e104cb First published 2022-03-02 Last updated 2023-06-26 ## Files and Weights 7 files, 673.3 MB in total. The weights are 2 files totalling 672.8 MB in bin, safetensors. Weights2 files · 672.8 MB Configuration1 file · 467 B Tokenizer1 file · 378.0 KB Documentation1 file · 1.9 KB Other1 file · 107.4 KB Repository1 file · 1.2 KB Every file | File | Type | Size | SHA-256 | | --- | --- | --- | --- | | model.safetensors | Weights | 336.4 MB | cbc89dbadf58 | | pytorch_model.bin | Weights | 336.4 MB | 08a1be6db147 | | config.json | Configuration | 467 B | — | | README.md | Documentation | 1.9 KB | — | | fill-mask-result.png | Other | 107.4 KB | — | | .gitattributes | Repository | 1.2 KB | — | | vocab.txt | Tokenizer | 378.0 KB | — | ## License and Download License apache-2.0 Access Open weights, no gate Download size 672.8 MB [Download from Kun](https://huggingface.co/jackaduma/SecBERT) Released by Kun through its official repository on Hugging Face. [Read the license](https://www.apache.org/licenses/LICENSE-2.0). ## Built From - Trained on (disclosed) APTnotes - Trained on (disclosed) CASIE - Trained on (disclosed) Stucco-Data ## Memory Requirements | Precision | Weights in memory | | --- | --- | | As published | 672.8 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. ## Compare SecBERT - [distilbert-base-uncased vs SecBERT](https://savrn.com/models/compare/distilbert-base-uncased-vs-secbert) ## Questions About SecBERT ### How much GPU memory does SecBERT need? About 0.2 GB at 16-bit and 0.1 GB at 4-bit: the weights (84M parameters) plus a working margin. A long context needs more. ### What is the cheapest GPU to run SecBERT 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 SecBERT commercially? Yes. SecBERT is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors. ### What is SecBERT's context length? 514 tokens, from the maximum position embeddings in its published configuration. ## Similar Models Model · Fill mask ### [distilroberta-base](https://savrn.com/models/distilroberta-base) [DistilBERT community](https://savrn.com/model-publishers/distilbert) This model is a distilled version of the RoBERTa-base model. It follows the same training procedure as DistilBERT. The code for the distillation process can be found here. The model has 6 layers, 768 dimension and 12 heads, totalizing 82M parameters (compared to 125M parameters for RoBERTa-base). On average DistilRoBERTa is twice as fast as Roberta-base. We encourage users of this model card to check out the RoBERTa-base model card to learn more about usage, limitations and potential biases. You can use the raw model for masked language modeling, but it's mostly intended to be fine-tuned on a downstream task. See the model hub to look for fine-tuned versions on a task that interests you.… Open weights apache-2.0 83M parameters 514 tokens transformers [View model](https://savrn.com/models/distilroberta-base) Model · Fill mask ### [my_awesome_eli5_mlm_model](https://savrn.com/models/my-awesome-eli5-mlm-model) [Ibrahim Samir Elmansoury](https://savrn.com/model-publishers/ibrahimmansoury) This model is a fine-tuned version of distilbert/distilroberta-base on an unknown dataset. It achieves the following results on the evaluation set: The following hyperparameters were used during training: - learningrate: 2e-05 - trainbatchsize: 8 - evalbatchsize: 8 - lrschedulertype: linear - numepochs: 3 - Transformers 5.17.0 - Pytorch 2.11.0+cu130 - Datasets 4.8.5 - Tokenizers 0.23.2 Open weights apache-2.0 82M parameters 514 tokens transformers [View model](https://savrn.com/models/my-awesome-eli5-mlm-model) Model · Fill mask ### [distilroberta-base-finetuned-wikitext2](https://savrn.com/models/distilroberta-base-finetuned-wikitext2) [Pufhv98w](https://savrn.com/model-publishers/3809ksdfh) This model is a fine-tuned version of distilroberta-base on an unknown dataset. It achieves the following results on the evaluation set: The following hyperparameters were used during training: - learningrate: 2e-05 - trainbatchsize: 8 - evalbatchsize: 8 - lrschedulertype: linear - numepochs: 3.0 - Transformers 5.18.0 - Pytorch 2.11.0+cu130 - Datasets 4.8.5 - Tokenizers 0.23.2 Open weights apache-2.0 82M parameters 514 tokens transformers [View model](https://savrn.com/models/distilroberta-base-finetuned-wikitext2) Model · Fill mask ### [hEVO-base-o](https://savrn.com/models/hevo-base-o) [Mulin Lab](https://savrn.com/model-publishers/mulinlab) hEVO-base-o is the base checkpoint in the hEvo public-data model family. It was independently pretrained on 1000 Genomes Project and Human Genome Diversity Project sequences using self-supervised masked language modeling. The checkpoint provenance records no target benchmark label supervision. Use a Python environment with PyTorch and the checkpoint's tested Transformers version: Access is gated for peer reviewers and requires manual approval by MulinLab. Sign in to Hugging Face, submit an access request on this page, and wait for approval. After approval, run hf auth login with your own read token; token=True below uses your locally stored credential. The model and tokenizer load… Access requested at publisher 87M parameters transformers [View model](https://savrn.com/models/hevo-base-o) Model · Fill mask ### [distilbert-base-uncased](https://savrn.com/models/distilbert-base-uncased) [DistilBERT community](https://savrn.com/model-publishers/distilbert) This model is a distilled version of the BERT base model. It was introduced in this paper. The code for the distillation process can be found here. This model is uncased: it does not make a difference between english and English. DistilBERT is a transformers model, smaller and faster than BERT, which was pretrained on the same corpus in a self-supervised fashion, using the BERT base model as a teacher. This means it was pretrained on the raw texts only, with no humans labelling them in any way (which is why it can use lots of publicly available data) with an automatic process to generate inputs and labels from those texts using the BERT base model. More precisely, it was pretrained… Open weights apache-2.0 67M parameters 512 tokens transformers [View model](https://savrn.com/models/distilbert-base-uncased) Model · Fill mask ### [bert-base-chinese](https://savrn.com/models/bert-base-chinese) [BERT community](https://savrn.com/model-publishers/google-bert) This model has been pre-trained for Chinese, training and random input masking has been applied independently to word pieces (as in the original BERT paper). This model can be used for masked language modeling 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)). typevocabsize: 2 vocabsize: 21128 numhiddenlayers: 12 Open weights apache-2.0 103M parameters 512 tokens transformers [View model](https://savrn.com/models/bert-base-chinese) ## Kun [All models and datasets](https://savrn.com/model-publishers/jackaduma) ## Versions - [7c603df5bc4c](https://savrn.com/models/secbert/versions/7c603df5bc4c) · current 2026-09-18 ## Explore More - [All fill mask models](https://savrn.com/models/tasks/fill-mask) - [All models under apache-2.0](https://savrn.com/models/licenses/apache-2-0) - [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/jackaduma/SecBERT) - [How the hub is built](https://savrn.com/model-hub/methodology)