Skip to main content

Module mamba2

Module mamba2 

Source
Expand description

Mamba-2: Structured State Space Duality (SSD).

§Mamba-2

Structured State Space Duality (SSD). Mamba-2 recasts the selective SSM recurrence as a chunkwise algorithm built from batched GEMMs, making it tensor-core friendly for training while remaining exactly equivalent to the recurrent form used for decoding. See mamba2 for the full SSD math.

  • mamba2 — the SSD block. The residual layer stack, the full language model, and the bidirectional wrappers are the family-generic types in [crate::generic] (MambaLatentNet / MambaVocabNet / MambaBidiLayers).
  • [cache] — the conv-window + SSM-state carried between calls.
  • [ssd] — the pluggable chunkwise SSD algorithms (Minimal / Serial / SerialRecalculated) and the backend extension trait.

Modules§

cache
Mamba-2 Inference Caches
mamba2
Mamba-2 SSM Block — Structured State Space Duality (SSD)
prelude
Public re-exports for Mamba-2.
ssd
Mamba-2 chunkwise SSD algorithms