220B MoE with full 262K context runs on RTX 3090s: 251 tok/s on 4 cards

QuixiAI · x · 2026-11-20

Developer superalesha built maya-hy4f, an engine on Project Maya kernels that runs Tencent's Hy4-flash — a 220B MoE (512 experts/layer, 16 active) with full 262K context, vision, and MTP speculative decoding — on consumer hardware, now uncensored.

No pruning: 1/2-card outputs agree with 4-card runs as much as two 4-card runs agree with each other.

Original post →

More from Infra

Infra channel →