NTv3 genome foundation model released with 15-species function prediction and regulatory sequence design
anshulkundaje · x · 2026-10-30
- InstaDeep released Nucleotide Transformer v3 (NTv3), pre-trained on OpenGenome2's 128k+ genomes (8 trillion nucleotides) spanning bacteria, archaea, eukaryotes and viruses, with basepair-resolution predictions on 1Mb sequences.
- Three post-training tracks: multi-task regression over 16,262 functional tracks across 15 species (expression, histone marks, TF binding, accessibility); gene-annotation finetuning on 174 assemblies producing models for mammals, plants and insects with HMM decoding; and masked-diffusion generative training on MPRA data for customized promoter/enhancer design.
- A revised preprint with full details accompanies the release.
More from Research
- Dev ditches unanimous-vote stopping rule for a judge in multi-AI debate project — lluisantoni · 2026-10-30
- Google's AI Search Uses Your Knowledge Graph Facts Without Linking, RESONEO Research Finds — dejanseo · 2026-10-30
- Raon-1-mini-preview runs natively on hipfire but 200x slower prefill than llama.cpp, patch released — Queasy_Asparagus69 · 2026-10-30
- TEM: Transformer Energy-based Models scale beyond diffusion without guidance — du_yilun · 2026-10-30
- NVIDIA's AgentBaton cuts agent API costs 65.9% with a single model handoff — dair_ai · 2026-10-30
- Stanford releases RT-J, a System 1 model predicting structured data in one forward pass — jure · 2026-10-30