[Quantization] Enable shared expert fusion compatibility with online shared_expert quantization (showcase: along Quark MXFP4 routed experts) (#55686)

Signed-off-by: Felix Marty <Felix.Marty@amd.com>
Signed-off-by: fxmarty-amd <felmarty@amd.com>
Co-authored-by: OpenAI Codex <codex@openai.com>
Co-authored-by: Misha Goin <mgoin64@gmail.com>
This commit is contained in:
fxmarty-amd
2026-10-08 12:04:46 -04:00
committed by GitHub
co-authored by OpenAI Codex Misha Goin
parent e2414ec1db
commit 9f3d70e081
24 changed files with 984 additions and 73 deletions
@@ -528,6 +528,7 @@ def test_padded_moe_reload_releases_each_layer(
layer.quant_method = method
layer.expert_map_manager = SimpleNamespace(map_global_to_local=lambda i: i)
layer._loaded_expert_biases = set()
layer._fused_shared_expert_quantizer = None
method.create_weights(
layer,
experts,