Single-cell Models
Single-cell foundation models are trained on millions of transcriptomic profiles to learn gene expression programs that define cell identity, state, and trajectory across tissues and conditions. By capturing the high-dimensional relationships between genes in individual cells, they enable cell type annotation, perturbation response prediction, and the discovery of rare cell populations without manual feature engineering. These models are accelerating our understanding of development, disease heterogeneity, and the cellular basis of complex traits.
42 models in this category
Notable Models
Top-rated single-cell models from our evaluations
GREmLN
Chan Zuckerberg Initiative / Columbia University / Chan Zuckerberg Biohub
A graph-signal-processing foundation model that embeds gene regulatory network structure directly into its attention mechanism for parameter-efficient single-cell transcriptomics.
mLLMCelltype
Texas A&M University
Multi-LLM consensus framework for automated cell type annotation in scRNA-seq data, outperforming prior methods by ~15% in mean accuracy.
scLDM.CD4
Chan Zuckerberg Initiative
A fine-tuned scLDM variant trained on 14.5 million CD4+ T cells for counterfactual prediction of single-gene perturbation effects in immune cells.
scLDM
Chan Zuckerberg Initiative
A scalable latent diffusion model for generating realistic single-cell gene expression profiles, using a permutation-invariant VAE and flow-matching diffusion transformer.
scPRINT
Institut Pasteur / CNRS
Foundation model pre-trained on 50 million single cells for robust gene network inference, with zero-shot denoising, batch correction, and cell type prediction.
STATE
Arc Institute
Transformer model for predicting cellular responses to perturbations across diverse cell contexts, trained on over 267 million human single-cell profiles.