Skip to content

vllm.models.mistral.nvidia.ops

Hand-fused eager kernels for Mistral models.

VLLM_ML4_FUSIONS selects them: a comma-separated subset of FUSIONS, all (default) or none.

Modules:

  • indexer –

    DSA indexer Q/K epilogues.

  • layernorm –

    One-launch LayerNorm: fp32 statistics and affine, output in the input dtype.

  • ssmax –

    SSMax query scaling q *= s * log(pos + 1) + b, alone or fused with RoPE.