DeepSeek-V4.1-Flash prefill on a single RTX 5090: 1,400 → 1,950 t/s (+39%) 🚀
One idea: stop re-decoding busy experts' weights 3–6× per chunk -decode once to FP16, let cuBLAS do the math.
Shipped it upstream to the Strata repo → PR #
4366#, keen to hear what you think
@coldniko