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 module

mla_prefill_sparse

SM100 (B200) sparse MLA prefill kernel with BF16 KV cache.

BF16-KV variant. K/V latents are read as BF16 via SWIZZLE_128B gather4 TMA and fed directly to the shared QK/SV MMA pipeline. The dtype-agnostic machinery (Q-load prologue, mma, cp_q_from_smem_to_tmem, _raw_indices_to_tma_rows, kv_valid_producer) is shared with the FP8-KV variant (mla_prefill_sparse_kv_fp8.mojo) through MLAPrefillSparseCommon in mla_prefill_sparse_utils.mojo.

Structs

• ​MLAPrefillSparse:

Functions

• ​mla_prefill_sparse:

Was this page helpful?

Thank you! We'll create more content like this.

Thank you for helping us improve!

😔 What went wrong?

Some code doesn’t work

It includes inaccurate information

It's missing information I need

It was difficult to understand

Other

Submit