JordiSilvestre/Spectral-AI

"O(log N) MoE Expert routing via RT Core ray tracing. BVH traversal replaces matrix multiplication in neural language models."

2 best points · 0 comments · 1 Hacker News post · 84 GitHub stars · View on GitHub

Hacker News discussions

• Repurposed Nvidia RT Cores for LLM routing (218x speedup) (2 points, 0 comments, by Jordisilvestre, 2026-04-09)