All Competitors

Every biological foundation model, evaluated and ranked by the bio.rodeo team

Showing 18 of 8 filtered models

  • CellOS

    Vitaura +2 othersJune 23, 2026batch_integrationcell_type_annotationfoundation_model+6

    Multi-view single-cell foundation model at 12B parameters, aligning expression and perception views with an LLM-JEPA joint-embedding objective.

    Single-cell
    8Openness
  • scLong

    2310
    Chinese Academy of SciencesApril 1, 2026batch_integrationcell_type_annotationfoundation_model+5

    Billion-parameter single-cell foundation model with self-attention over 28,000 human genes, adding Gene Ontology priors via a graph neural network.

    Single-cell
    29Openness
  • RegFormer

    BGI ResearchApril 1, 2026cell_clusteringbatch_integrationperturbation_modeling+7

    Single-cell foundation model combining regulatory network priors with a Mamba backbone for clustering, batch integration, and perturbation modeling.

    Single-cell
    10Openness
  • scDiVa

    1
    Renmin University of ChinaFebruary 3, 2026gene_expressioncell_type_annotationperturbation_prediction+6

    Single-cell foundation model built on masked discrete diffusion, jointly generating gene identities and expression values from 59 million cells.

    Single-cell
    6Openness
  • University of PennsylvaniaOctober 1, 2025batch_integrationcell_type_annotationfoundation_model+4

    Kidney-specialized single-cell foundation model trained across four mammalian species for zero-shot cell-type annotation and batch integration.

    Single-cellSpatial omics
    22Openness
  • scDMC

    2
    Yunnan UniversitySeptember 11, 2025batch_integrationcell_type_annotationcontrastive_learning+2

    Single-cell foundation model with rank and expression-aware input streams, pairing masked gene modeling with cell-level contrastive learning.

    Single-cell
    26Openness
  • New Jersey Institute of TechnologyApril 22, 2025batch_integrationcell_type_annotationcontrastive_learning+2

    Single-cell foundation model built on bidirectional Mamba blocks and pretrained on 30 million cells for linear-time transcriptome embedding.

    Single-cell
    67Openness
  • Brown UniversityApril 16, 2025batch_integrationcontrastive_learningmultilayer_perceptron+2

    Single-cell RNA-seq encoder trained with contrastive learning to merge plate- and droplet-based protocols, zero-shot on unseen tissues.

    Single-cell
    39Openness