[CPU] Refactor CPU WNA16 (#28826)

Signed-off-by: jiang1.li <jiang1.li@intel.com>
2025-11-19 10:32:00 +08:00
parent 40b6b38f2c
commit 20852c8f4c
22 changed files with 1656 additions and 78 deletions
--- a/vllm/config/model.py
+++ b/vllm/config/model.py
@@ -1020,6 +1020,8 @@ class ModelConfig:
                # Ensure heavy backends are probed last to avoid unnecessary
                # imports during override detection (e.g., MXFP4 imports Triton)
                "mxfp4",
+                "cpu_gptq",
+                "cpu_awq",
            ]
            quantization_methods = [
                q for q in supported_quantization if q not in overrides