All Competitors

De novo protein design from natural language: a 16B-parameter framework turning text descriptions into sequences via structure-conditioned generation.

General medical vision-language model trained with reinforcement learning to reason step by step over medical images for diagnosis and visual QA.

Hypergraph foundation model for brain disease diagnosis from resting-state fMRI, self-supervised on high-order connectivity among brain regions.

265M-parameter genomic foundation model whose BioToken scheme folds variants and region annotations into the tokens for variant effect prediction.

Open therapeutics foundation models from Google, built on Gemma-2, for drug-discovery property prediction and conversational reasoning.

Protein dynamics model predicting per-residue probability of microsecond-millisecond conformational exchange from sequence or structure.

Medical vision-language model trained with reinforcement learning for generalizable reasoning across eight imaging modalities and five question types.

Genomic language models with disentangled attention, pretrained on prokaryotic and eukaryotic genomes for sequence classification and variant effects.

Histopathology foundation model pretrained with BEiT masked image modeling on 11M+ tissue image tiles for cancer diagnosis and survival prediction.

Multimodal LLM unifying 12-lead ECG time series, ECG images, and text for grounded, clinician-aligned electrocardiogram interpretation.

Protein structure embedding model that compresses each 3D fold into a single fixed-length vector for proteome-wide similarity search and clustering.

Single-cell RNA-seq foundation models combining masked modeling with ontology supervision to classify cell states across unseen donors and diseases.

Patch-based 3D diffusion model that generates teravoxel-scale virtual mouse brain volumes conditioned on spatially resolved gene expression.

Protein sequence design model that represents small molecules, nucleotides, and metals at atomic resolution, enabling ligand-aware enzyme design.

Universal ultrasound segmentation foundation model adapting the Segment Anything Model to eight anatomical regions in a single promptable network.

Latent diffusion model for single-cell multi-omics generation and modality translation, with gradient-based inference of gene regulatory networks.

2B-parameter medical vision-language model that uses reinforcement learning to show interpretable reasoning for radiology visual question answering.

Whole-brain axon and soma segmentation foundation model and registration pipeline for developmental connectomics, with no per-stage retraining.

Brain-to-text decoder that reconstructs typed sentences from non-invasive MEG and EEG brain recordings using a CNN, transformer, and language model.

Single-cell DNA methylation foundation model capturing genome-wide CpG dependencies in whole-genome bisulfite sequencing across tissues and species.

Histology-anchored framework pairing an H&E foundation model with a cellular hypergraph to predict single-cell multi-omics from tissue images.

Genomic foundation model trained on 9.3 trillion DNA base pairs across all domains of life, with 40B parameters and a 1-million-token context.

Vision-language foundation model for fetal ultrasound, pretrained on 210,035 image-text pairs for plane classification, biometry, and segmentation.

Chest X-ray vision-language model that drafts the findings section of a radiology report, at 7B parameters small enough to run on a single GPU.