Non-autoregressive System 1 decision engine. Typed choice, score and yes/no decisions over any text in a single forward pass, in 100+ languages, with a router that picks the right checkpoint per request.
Python calibrationclassificationdecision-modelhuggingfacejevmodernbertmultilingualnlppythonpytorchroutingtyped-decisionszero-shot
perf(fast): reduce GEGLU pipeline stages
周诚宇(Zhou Chengyu)c9da251e01f1
diff --git a/laya/tl_kernels.py b/laya/tl_kernels.pyindex b5c9ee61..8affc84a 100644--- a/laya/tl_kernels.py+++ b/laya/tl_kernels.py@@ -56,7 +56,7 @@ def gemm_kernel(N, K, bias=False, act="none", bm=64, bn=128, bk=64, stages=3, th@tilelang.jit(pass_configs=FAST)-def gemm_geglu_kernel(F, K, bm=64, bn=64, bk=64, stages=3, threads=128, dtype="bfloat16"):+def gemm_geglu_kernel(F, K, bm=64, bn=64, bk=64, stages=2, threads=128, dtype="bfloat16"):"""ModernBERT GLU MLP up-projection, fused: C[M,F] = gelu(A @ Wi[:F]^T) * (A @ Wi[F:]^T)."""DT = _dt(dtype)M = T.dynamic("M")