[Quantization][CI] add humming lm-eval test (#43752)

Signed-off-by: Jinzhen Lin <jinzhen.ljz@antgroup.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
This commit is contained in:
Jinzhen Lin
2026-06-22 11:23:55 -07:00
committed by GitHub
co-authored by mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
parent 3e6529cc0e
commit e4b3da3feb
9 changed files with 91 additions and 2 deletions
@@ -852,7 +852,7 @@ class HummingMoEMethod(FusedMoEMethodBase):
# use moe modular
experts: HummingIndexedExperts | HummingGroupedExperts
layer.ensure_moe_quant_config_init()
layer._ensure_moe_quant_config_init()
assert self.moe_quant_config is not None
if get_humming_moe_gemm_type() == "indexed":
experts = HummingIndexedExperts(layer, self.moe, self.moe_quant_config)