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

topk_bitonic

Block-wide bitonic sort top-k for the MLA/MSA indexer.

This is to replace the two-stage sequential-extraction topk_gpu is pathological when k ≈ N (near-full sort). A use case of k = N = 2048 is the DeepSeek-V3 / MiniMax-M3 indexer config.

comptime values

PERSISTENT_TOPK_MAX_N

comptime PERSISTENT_TOPK_MAX_N = Int(2048)

Functions

• ​persistent_topk_block: Launch block-wide bitonic sort top-k for total_seq_len score rows.

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