Geneformer

Single-cell transcriptome foundation model for context-aware network-biology predictions, from the Broad Institute / Theodoris lab.

Unsigned

Overview

Geneformer is a foundational transformer model pretrained on a large-scale corpus of human single-cell transcriptomes (Genecorpus-30M, and now Genecorpus-104M for the V2 model), developed by the Theodoris lab at the Broad Institute. It enables context-aware predictions in network biology settings with limited data, including zero-shot in-silico perturbation and gene/cell-state classification, and is also listed on the Chan Zuckerberg Initiative's Virtual Cells Platform.

**Registry note on `model_card_uri`**: this points to the model's standard Hugging Face model card (its README carries a `license: apache-2.0` YAML tag, confirmed directly), not a GA4GH Genomic AI Model Card v0.6.0-formatted document. No model has yet published a card in the GA4GH v0.6.0 format, since it is a newly drafted specification.

Details

Licence: Apache-2.0
Version: V2 (Genecorpus-104M)
Category: Genomic foundation model
Homepage: https://huggingface.co/ctheodoris/Geneformer
Repository: https://huggingface.co/ctheodoris/Geneformer
Maintainers: Susheel Varma (@susheel, GA4GH AI Workstream / Sage Bionetworks)
Keywords: single-celltranscriptomicsgenomicsfoundation-modelnetwork-biology

Safety classification

No safety classification has been submitted for this entry.

Model Card

View the full GA4GH Genomic AI Model Card ↗

Hosting: huggingface
Access: https://huggingface.co/ctheodoris/Geneformer

Mirrored data last synced from the Model Card at 2026-09-04T11:05:00.000Z. See the linked card for the authoritative, live version.

Provenance

Created: 2026-09-04T11:05:00.000Z
Updated: 2026-09-04T11:05:00.000Z
Last verified: 2026-09-04T11:05:00.000Z