Cherimoya: A Model Tailored for Genomic Data
jmschreiber91 · x · 2026-07-15
Cherimoya is a model designed for genomic data, built upon the Cheri block, which itself is an adaptation of the ConvNeXT architecture for genomic applications.
To enhance speed, the team implemented two custom GPU kernels:
- One for training
- A megakernel for inference
The post emphasizes that the project goes beyond architectural adaptation, offering specialized engineering optimizations for both training and inference.
Related event: Cherimoya: A Small, Fast Genomic S2F Model Reaching SOTA(6 posts)→
More from Research
- Nat Lambert shares a reading list on synthetic data and agentic SFT data — natolambert · 2026-07-22
- Lightwheel AI Launches SimReadyGen: Text-to-Physics-Accurate Robot Sim Assets — ZeYanjie · 2026-07-22
- PNAS special issue examines copyright, governance, and AI in the legal system — chrmanning · 2026-07-22
- WeirdChat catalogs strange model behaviors from more than 100 million sampled responses — JacobSteinhardt · 2026-07-22
- New agentic benchmark shows AI managers escalate to coercion and fake success — Jasmine Brazilek · 2026-07-22
- Ai2’s Asta adds one-click handoff and self-checking deep paper search — allen_ai · 2026-07-22