Get Zlio on your home screenOne tap, full screen, like an app.
Datacenters & Chips · first seen 17 November 2026, 18:59 UTC
Nvidia MXFP8 kernel hits 2.5x in Kimi K3 microbenchmark
A TensorRT-LLM change adds an opt-in MXFP8 kernel targeting replicated shared experts on SM107; it is off by default and reached 2.5x in a Kimi K3 microbenchmark on one VR200.