HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks

deberta-v3-base

microsoft/deberta-v3-base

kind
original
quant
bf16 (16-bit)
family
misc
modality
fill-mask
root model
microsoft/deberta-v3-base
provider
microsoft
license
mit
language
—
context
—
library
—
vRAM (GB)
—
size on disk
—
downloads
2,700,642
likes
430
created
—
last modified
2022-09-22T12:34:19+00:00 (2022)
huggingface
https://huggingface.co/microsoft/deberta-v3-base
tags
transformerspytorchtfrustdeberta-v2debertadeberta-v3fill-maskenarxiv:2006.03654arxiv:2111.09543license:mitendpoints_compatibleregion:us

Benchmark scores (0)

no scores extracted for this variant.

Discussions (18)

titleauthorstatuscommentstype
Listed on OpenModelMapduola15open0disc
TemporalMesh Transformer: 29.4 PPL at 48% compute — beats Mamba, new open-source architecturevigneshwar234open0disc
fine-tuning the deberta on token length greater than 512iamhdaveopen0disc
Reflective Understanding in Language Modelselly99open0disc
Adding `safetensors` variant of this modelSFconvertbotopen0disc
deberta base generator model rquesthd2323open0disc
Deberta max lengthdataminer1open0disc
added license file based on tagiamankitopen0disc
Fix typo in README.mdnelsonauneropen0disc
Inference timeavifaizaopen0disc
Bizzare responsesumarbutleropen0disc
Add model_max_length to tokenizercoreyhuclosed0disc
[AUTOMATED] Model Memory Requirementsmodel-sizer-botopen0disc
Examples yield bad resultsxorimaopen0disc
Adding `safetensors` variant of this modelSFconvertbotopen0disc
Generator Modelprajwal967open0disc
Adding ONNX file of this modelcouturiercopen0disc
Add task to enable widgetlewtunmerged0disc
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.