Evo 2 generates complete Cas9 loci from 4 natural loci in prompt at ~10x less compute
garykbrixi · x · 2026-10-29
The author reports that on protein-RNA systems, neither Evo 1 nor Evo 2 could generate Cas9 from partial loci, and producing EvoCas9 required fine-tuning on tens of thousands of loci.
With four natural loci in the prompt, Evo 2 generates complete loci containing a CRISPR array, tracrRNA and Cas9. Per-sample pass rate is lower than fine-tuning, but generating a library costs about 10x less compute and yields more diverse designs.
More from Research
- Stanford releases RT-J, a System 1 model predicting structured data in one forward pass — jure · 2026-10-30
- Generalist AI ships GEN-1.5 and challenges the community to SLAM its wrist-cam data — E0M · 2026-10-30
- Signal65 Launches PINNACLE Coding Agent Benchmark: One Model Aces 97% of Samples, Fails All 30 Jobs — ryanshrout · 2026-10-30
- TEM: Transformer Energy-Based Models Claim to Outscale Diffusion Without Guidance — du_yilun · 2026-10-30
- VisualForce: robots infer grip force from soft gripper deformation, no force sensors needed — du_yilun · 2026-10-30
- Microsoft Research: coding agents' real bottleneck is understanding code, not writing it — Microsoft Research · 2026-10-30