Jade: GPU-Accelerated Homology Search Engine, 26x Faster Than BLASTN
anshulkundaje · x · 2026-11-15
Litefold released Jade, a lightweight GPU-accelerated homology search engine for proteins, DNA, and RNA.
Modern protein structure prediction relies heavily on evolutionary information from MSAs, but generating them at scale demands large databases, memory, and preprocessing infrastructure.
Key details:
- Built on DIAMOND, adding custom CUDA kernels, nucleotide search, and direct A3M generation for co-folding models like Boltz-2 and Protenix — no prebuilt search index required
Benchmarks:
- 26x faster than BLASTN at near-matched recall on simulated divergent DNA sequences
- 10–11x faster protein search on an A100 vs a 64-thread CPU baseline, with unchanged sensitivity
- 49 seconds to prepare all 32 MSAs across 15 CASP17 targets, with MSA prep at only 3–5% of end-to-end co-folding time
More from Infra
- CoreWeave says GPU pricing jumped 30-50% QoQ, above original contract rates — Beth_Kindig · 2026-11-15
- Americans: no data centers in our backyard! SpaceX: no problem — draecomino · 2026-11-15
- The Starship math: 25,000 flights to deploy 250GW of space compute — JOBhakdi · 2026-11-15
- Microsoft patent would pay gamers to turn idle gaming GPUs into AI training hardware — Shot-Height-7194 · 2026-11-15
- InferSSD streams MoE experts from SSD so you can run models bigger than your VRAM — AIFlow_ML · 2026-11-15
- Pre-training compute share collapses from 67% to 7% as inference and RL take over — robleclerc · 2026-11-15