nvmath.bindings.cublas.gemm_strided_batched_ex

nvmath.bindings.cublas.gemm_strided_batched_ex(intptr_t handle, int transa, int transb, int m, int n, int k, intptr_t alpha, intptr_t a, int atype, int lda, long long stride_a, intptr_t b, int btype, int ldb, long long stride_b, intptr_t beta, intptr_t c, int ctype, int ldc, long long stride_c, int batch_count, int compute_type, int algo)

See cublasGemmStridedBatchedEx.