Genomic-NIAH: needle-in-a-haystack long-context retrieval benchmark for DNA language models
- Provider
- Hugging Face (HuggingFaceBio, Carbon team)
- License
- Apache-2.0 (source)
- Size
- 28 files
- Format
- PNG, Parquet (in sampled files)
- Stored at
- Hugging Face (https://huggingface.co/datasets/HuggingFaceBio/genomic-niah)
- Access
- Free and public. No account with the data provider is needed.
- Category
- Genomics & Bioinformatics
- Source
- https://huggingface.co/datasets/HuggingFaceBio/genomic-niah
About this dataset
A long-context retrieval benchmark for genomic language models from Hugging Face, built for the Carbon DNA models. Each example plants a random 24 bp KEY and 24 bp VALUE pair inside a real eukaryotic genome haystack and asks the model to return VALUE given the haystack followed by KEY, so success requires actual long-range retrieval. Four tasks (plain retrieval and three near-duplicate variants with eight distractor keys differing by 4, 2 or 1 bp, inspired by RULER multi-key) cross six context lengths from 4k to 128k 6-mer tokens (about 24 kbp to 786 kbp), giving 24 configurations of 500 examples each (12,000 in total), stratified over five needle depths (10 to 90%) and four kingdoms (animalia, fungi, plantae, protista). Haystacks come from the OpenGenome2 NCBI eukaryotic test and validation splits, using contiguous ACGT runs up to 16k tokens and stitched within-record runs beyond. Data are Parquet under data/, one config per task and length (e.g. niah_16k, niah_neardup_d1_64k), loadable with Hugging Face datasets; about 4 GB. Released under Apache 2.0; cite the Carbon paper (Ben Allal et al., 2026).
How to load this dataset in Python
from datasets import load_dataset
ds = load_dataset("HuggingFaceBio/genomic-niah")
print(ds)Data structure
File types in sampled files: PNG, Parquet
Files
Example file: data/niah_128k/test-00000-of-00001.parquet
Analyze this dataset without downloading it
On Scigantic, Genomic-NIAH: needle-in-a-haystack long-context retrieval benchmark for DNA language models opens in a hosted JupyterLab notebook with the data already mounted, so you can read files where they are stored instead of copying them first. Each notebook includes an AI assistant that can see the mounted files and write and run analysis code. Open the dataset page to launch a notebook, or browse the full catalog.
This page is also available as Markdown and as JSON.
Related Genomics & Bioinformatics datasets
- Tiny OpenGenome2
- Data from: Genome assembly of Danaus chrysippus and comparison with the Monarch Danaus plexippus
- PromoterZoo: 13.6 Million Multi-Species Promoter Sequences for Training the LOL-EVE Variant Effect Model (Harvard Marks Lab)
- COVID-19 Genome Sequence Dataset
- Rye Reference NLRome
- TCGA-TGCT Testicular Germ Cell Tumors (GDC open-access tables, 263 cases)
- Reactome (curated human pathway database)
- Aging Gene Expression Single Cell Mouse