All Competitors

A reinforcement-learning-driven generative framework that unifies sequence generation with multi-objective RNA codon optimization, generalizing across six species and five RNA types without per-dataset retraining.

A promptable DNA language model that generates multi-kilobase plasmid sequences from human-readable component specifications, post-trained with verifiable rewards.

Multimodal reverse-translation language model that generates species-aware mRNA coding sequences from protein sequences, conditioned on host taxonomy.

Transformer that generates multi-species antibody and nanobody framework regions at the mRNA level, conditioned on input CDRs, across six species.

An mRNA language foundation model trained on ~115M coding sequences across the tree of life for unified mRNA sequence perception and generation.

GPT-2-style generative foundation model pretrained on ~165M mRNA coding sequences across all three domains of life for de novo CDS generation and design.