Tutorials

These tutorials introduce Ave through small kernels that showcase the language and runtime.

1. Tutorial 1: AXPY Introduce `@ave.jit`, typed tensors, thread indexing, and kernel launch geometry.
2. Tutorial 2: GEMM Start with naive BF16 `A @ B^T`, then add shared-memory tiling and vectorized tile loads.
3. Tutorial 3: Intrinsics Use AMDGPU MFMA, runtime pointer views, and raw-buffer loads for generic GEMM kernels.