For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /get-started.md).

Mojo package

builtin_kernels

Provides the graph compiler's built-in kernel registrations.

Each module binds graph op names (such as mo.matmul) to the open-source kernels in linalg, nn, comm, and related packages using the same @extensibility.register mechanism available to custom ops, so these registrations double as worked examples of how kernels connect to MAX graphs.

Modules

• ​attention: Registers attention graph ops (MHA, MLA, fused QKV) and dispatches them to the nn.attention kernels.
• ​attn_res: Graph-op binding for Kimi K3's attention-residual softmax mixture.
• ​conv: Registers convolution graph ops and dispatches them to the nn.conv kernels.
• ​distributed: Registers distributed and multi-GPU collective graph ops backed by the comm and shmem kernels.
• ​elementwise: Registers elementwise and unary/binary math graph ops.
• ​ep: Expert Parallelism (EP) Communication Kernel.
• ​gather_scatter: Registers gather, scatter, and indexed-access graph ops backed by the nn.gather_scatter kernels.
• ​kda: Graph-op bindings for the KDA (Kimi Delta Attention) kernels.
• ​kernels: Registers core tensor graph ops (range, copy, reshape, and related utilities).
• ​kv_cache: Registers KV-cache graph ops backed by the kv_cache and nn.kv_cache kernels.
• ​linalg: Registers matmul-family graph ops (matmul, grouped, batched, and quantized variants) and dispatches them to the linalg kernels.
• ​logprobs: Registers log-probability computation graph ops used by sampling pipelines.
• ​mega_ffn: Graph-op binding for the fused MegaFFN MoE FFN composite op (NVFP4 + MXFP8).
• ​msa: Graph-op bindings for the MiniMax-M3 block-sparse attention (MSA) kernels.
• ​mtp: Graph-op bindings for multi-token-prediction draft layers.
• ​nan_check: NaN/Inf detection kernels for the max-debug.nan-check feature.
• ​qsa: Graph-op bindings for the QSA lightning indexer and its attention consumer.
• ​quantization: Registers quantization and dequantization graph ops backed by the quantization and linalg kernels.
• ​reductions: Registers reduction graph ops (sum, mean, argmax, and related) over tensor axes.