Local benchmarks reveal Hunyuan Hy4 Flash speed and long-prompt load times

Testers benchmark Tencent's Hunyuan Hy4 Flash locally, reaching 45.5 tok/s on a single DGX Spark and 137 tok/s on dual Macs, while loading a 1M-token prompt takes 44 minutes on Mac Studio M3 Ultra and 81 minutes on DGX Spark.

2026-11-20 ~ 2026-11-21 · 2 related posts