mamba-rs 0.5.0

Mamba SSM and Mamba-3 SISO in Rust with optional CUDA GPU acceleration. Inference and training (BPTT through SSM state, AdamW), CPU + GPU paths, custom CUDA kernels, CUDA Graph capture, f32 / bf16 / f16. Opt-in deterministic training (bit-identical runs, batch-invariant inference) with a tensor-core tier that beats cuBLAS on LLM-sized models.
Documentation

mamba-rs

There is very little structured metadata to build this page from currently. You should check the main library docs, readme, or Cargo.toml in case the author documented the features in them.

This version has 7 feature flags, 0 of them enabled by default.

default

This feature flag does not enable additional features.

accelerate

This feature flag does not enable additional features.

cli

cuda

cudarc

gemm

gemm-blas

hf