Linear algebra

GEMM, factorizations and low-bit weight storage.

Cholesky	Cholesky factorization A + jitter * I = L @ L.T of a symmetric positive-definite matrix, then solves.
linalg	Float32 linear algebra: a bit-identical matrix product, low-bit products and NumPy-named decompositions.
lowbit	Low-bit weight storage for the inference classes, with float32 arithmetic.
matmul	Float32 matrix product op(a) @ op(b) with the same bits on NVIDIA and AMD.