YDXX/ViroHyena-253m

ViroHyena

ViroHyena is a Hyena-based nucleotide language model pre-trained on the ViroBland (ViroBlend) corpus, a small (216 Mbp) mixed pretraining dataset with source-wise stratified sampling to balance human reference, multi-species genomes, and viral in-domain sequences.

This model was introduced in the paper ViroBench: Benchmarking Nucleotide Foundation Models on Viral Genomics Tasks.

Code: GitHub - ViroBench

Model Configurations

Model	Params	d_model	Layers
ViroHyena-436K	0.436M	128	2
ViroHyena-1.6M	1.6M	256	2
ViroHyena-6.6M	6.6M	256	8
ViroHyena-253M	253M	1024	20

Citation

If you use this model in your research, please cite:

@article{ye2026virobench,
  title={ViroBench: Benchmarking Nucleotide Foundation Models on Viral Genomics Tasks},
  author={Ye, Dongxin and Hu, Fang and Hu, Han and Hu, Shu and Tan, Yang and Ouyang, Wanli and Li, Stan Z and Cui, Jie and Dong, Nanqing},
  journal={arXiv preprint arXiv:2605.25388},
  year={2026}
}