Skip to content

vllm.models.deepseek_v4.xpu

Modules:

  • dspark –

    DSpark draft model for DeepSeek-V4 on Intel XPU.

  • model –
  • mtp –

    MTP draft model for DeepSeek V4 (internal codename: DeepseekV4).

  • xpu_qnorm_rope_kv_fp8_insert –

    XPU Triton replacement for fused_deepseek_v4_qnorm_rope_kv_rope_quant_insert.

  • xpu_sparse –

    XPU DeepSeek-V4 attention subclass.

  • xpu_sparse_decode_fp8 –

    XPU sparse decode for DeepSeek V4 with FP8 KV cache.