For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /max/get-started.md).
Mojo function
matmul_dynamic_block_scaled_mxfp4
def matmul_dynamic_block_scaled_mxfp4[out_dtype: DType](c: TileTensor[out_dtype, address_space=c.address_space, linear_idx_type=c.linear_idx_type, element_size=c.element_size], a: TileTensor[DType.uint8, address_space=a.address_space, linear_idx_type=a.linear_idx_type, element_size=a.element_size], b: TileTensor[DType.uint8, address_space=b.address_space, linear_idx_type=b.linear_idx_type, element_size=b.element_size], a_scales: TileTensor[DType.float8_e8m0fnu, address_space=a_scales.address_space, linear_idx_type=a_scales.linear_idx_type, element_size=a_scales.element_size], b_scales: TileTensor[DType.float8_e8m0fnu, address_space=b_scales.address_space, linear_idx_type=b_scales.linear_idx_type, element_size=b_scales.element_size], ctx: DeviceContext)
Was this page helpful?
Thank you! We'll create more content like this.
Thank you for helping us improve!