[CPU] Optimize Qwen3-next model on CPU (#12525)

Co-authored-by: Ma Mingfei <mingfei.ma@intel.com> Co-authored-by: Fan Yin <1106310035@qq.com>
2026-01-30 14:03:58 +08:00
parent 71e4d3b6bc
commit 336dc4579e
13 changed files with 366 additions and 41 deletions
--- a/sgl-kernel/python/sgl_kernel/init.py
+++ b/sgl-kernel/python/sgl_kernel/init.py
@@ -78,7 +78,13 @@ from sgl_kernel.kvcacheio import (
    transfer_kv_per_layer,
    transfer_kv_per_layer_mla,
 )
-from sgl_kernel.mamba import causal_conv1d_fwd, causal_conv1d_update
+from sgl_kernel.mamba import (
+    causal_conv1d_fn_cpu,
+    causal_conv1d_fwd,
+    causal_conv1d_update,
+    causal_conv1d_update_cpu,
+    chunk_gated_delta_rule_cpu,
+)
 from sgl_kernel.marlin import (
    awq_marlin_moe_repack,
    awq_marlin_repack,