All Competitors

Multi-modal protein language model using the MSA evolutionary profile as a reasoning step between structure and sequence. 650M outperforms ESM-3 1.4B.

Surface-EMG wristband models that decode hand gestures, handwriting, and wrist movement, generalizing across users without per-person calibration.

Protein-ligand binding affinity and mutation ΔΔG predictor fusing residue, ligand, and interaction graphs, evaluated on leak-proof LP-PDBBind splits.

Tandem mass spectrum prediction for intact N- and O-glycopeptides that localizes O-glycosylation sites from HCD spectra alone, without ETD.

Protein language models trained on billions of natural and synthetic sequences for de novo design and zero-shot mutation-effect prediction.

Whole-genome epigenetic foundation models reading DNA methylation and sequence jointly to detect Alzheimer's and Parkinson's from plasma cfDNA.

Splicing variant effect prediction across 49 human tissues and 15 developmental stages, from four weeks post conception to adulthood.

Bacterial genomics foundation model reading whole genomes as ordered protein sequences. Predicts operons, gene essentiality, and phenotypic traits.

Species-conditioned codon language model that jointly reads 5'UTRs, coding sequence, and RNA secondary structure to design native-like genes.

Controllable DNA sequence design conditioned on cell type, transcription factor, or activity signal, in GPT- and BERT-style transformer variants.

Structure-based drug design model that inpaints a 3D ligand density into an empty protein pocket, then decodes those voxels into valid SMILES.

Nanomaterial-protein interaction prediction from protein sequence, structure, and experimental context that generalizes to unseen materials.

RNA language model built on the xLSTM recurrent architecture instead of a transformer backbone, pretrained on RNAcentral with 33.7M parameters.

Structure-based drug design model that generates 3D ligands inside a protein pocket by interpolating distribution parameters instead of samples.

Antibody inverse folding by discrete diffusion, sampling sequences that fold onto a given backbone. Generates varied CDRs at high sequence recovery.

Sparse autoencoder for blood-cell microscopy that decomposes hematology foundation model embeddings into expert-validated sub-cellular concepts.

Bacterial promoter annotation and expression prediction from a 1.8M-parameter transformer pretrained on 9M gammaproteobacterial regulatory regions.

Synthesizable 3D molecule generation that jointly samples building blocks, reactions, and atomic coordinates, returning a synthesis route per design.

Attention-based multiple instance learning heads for whole-slide pathology, pretrained on a 108-way pan-cancer slide classification task.

Scoring function for protein, nucleic acid, and small-molecule complexes that predicts binding affinity, ranks docked poses, and screens ligands.

Plant immune receptor-ligand classifier that scores MAMP epitope immunogenicity from sequence, reaching 73% accuracy on a held-out test set.

Spatial transcriptomics foundation model pairing gene-scale cell embeddings with an SE(2) Transformer over cell coordinates, pretrained on 88M cells.

Enzyme Commission number prediction that pools ESM Cambrian embeddings across unlabeled sequence homologs, scoring 0.788 F1 on full 4-digit EC.