All Competitors

Genomic language model from Radical Numerics with a 2 Mbp context window, built for zero-shot variant effect prediction and sequence design.

Diffusion-based generative RNA model for de novo sequence design, conditioned on function, RNA family, structure, or binding proteins.

Multiscale graph neural network for fixed-backbone protein binder sequence design with a contrastive decoding algorithm to improve target selectivity.

RNA inverse-folding language model that designs nucleotide sequences satisfying a target secondary structure, fixed bases, and coding constraints.

110M-parameter RNA language model that designs sequences from secondary structure, motif, and Gene Ontology constraints via discrete diffusion.

RNA inverse folding framework pairing a graph neural network predictor with a diffusion model, designing sequences from self-contained RNA units.

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

DNA foundation model scaling an encoder-only transformer to 7 billion parameters for variant effect prediction, gene expression, and sequence design.

RNA foundation model with 1.6 billion parameters, pretrained on 42 million non-coding RNA sequences for structure prediction and RNA sequence design.

Genomic foundation model with 7B parameters that models prokaryotic DNA, RNA, and protein at single-nucleotide resolution over a 131k-token context.

Biochemistry-aware inverse folding model that augments backbone geometry with physicochemical point clouds, reaching ~90% sequence recovery on CATH.

Transformer language model for 5' UTR sequences that predicts mRNA translation efficiency, ribosome loading, and protein expression levels.

Generative RNA design model that samples family sequences from a VAE latent space constrained by Rfam covariance models and consensus structure.

Zero-shot antibody affinity maturation using ESM pseudolikelihood scoring. Improves binding up to 160-fold with no antigen-specific training data.

Message passing neural network for fixed-backbone protein sequence design. Achieves 52.4% native sequence recovery, far surpassing Rosetta's 32.9%.