baracuda-kernels-sys 0.0.1-alpha.78

Compiled bespoke .cu kernel template instantiations for the baracuda ML kernel facade plus C-ABI FFI facades for the library-backed plans (cuDNN conv/pool, cuSOLVER linalg, cuFFT/cuRAND, CUTLASS GEMM re-export). Hosts curated CUDA kernel sources (int8/FP8/int4/bin GEMM RRR, elementwise, reduce, norm, attention, …), builds them via baracuda-forge, exposes extern "C" entry points for the safe baracuda-kernels crate. CUTLASS template kernels live in the sibling baracuda-cutlass-kernels-sys crate and are re-exported here under the unified baracuda_kernels_gemm_* namespace.
Documentation

baracuda-kernels-sys

There is currently very little information to present on this page because Docs.rs has only limited support for extracting structured feature metadata from Cargo crates. This issue is tracked in Rust RFC #3416. Check this library's main docs, readme, and Cargo.toml in case its authors have documentation for features available there instead.

This version has 15 feature flags, 1 of them enabled by default.

default

sm80 (default)

awq

bnb_nf4

cudnn

This feature flag does not enable additional features.

fa2

fa2_backward

flashinfer

mamba

marlin

mhc

ring_attention

sm89

This feature flag does not enable additional features.

sm90a

xformers_blocksparse

xformers_sparse24