flashinfer.mla.nvfp4_quantize_append_sparse_mla_cache

flashinfer.mla.nvfp4_quantize_append_sparse_mla_cache(latent_kv: Tensor, slot_mapping: Tensor, cache: Tensor) → None

Quantize and append DeepSeek-V4 latent KV by physical cache slot.

Parameters:

  • latent_kv (torch.Tensor) – Contiguous CUDA BF16 tensor with one 512-element latent-KV row per entry in slot_mapping.

  • slot_mapping (torch.Tensor) – Contiguous 1D CUDA int32 or int64 tensor. slot_mapping[i] is page_id * page_size + entry_id. Negative and out-of-range slots are padding and are ignored. If a valid slot occurs more than once, the lowest-index input row is written deterministically.

  • cache (torch.Tensor) – Destination opaque uint8 paged cache. The 3D shorthand [num_pages, page_size, 384] and public 4D HND/NHD layouts are accepted. Page-strided views may place the cache inside vLLM’s packed physical block allocation. Only addressed data rows and scale slots are written, so prefill history is reused directly by decode.