vllm.models.kimi_k3.nvidia.ops ¶
Modules:
-
attn_res– -
cute_dsl– -
fused_mla_key_concat_kv_cache–Fused MLA prefill and decode epilogues for Kimi-K3.
-
latent_moe_tail– -
third_party– -
vision_fa4_warmup–Startup compilation of Kimi-K3 vision FA4 kernels.