[Model] add GLM-5.3-Flash support (#53906)

This commit is contained in:
Jiangyun Zhu
2026-09-03 09:40:35 -07:00
committed by GitHub
parent da8ec28268
commit 98ed0856f3
94 changed files with 17927 additions and 266 deletions
+1 -1
View File
@@ -142,7 +142,7 @@ def can_initialize(
patch.object(V1EngineCore, "_initialize_kv_caches", _initialize_kv_caches_v1),
monkeypatch.context() as m,
):
if requires_spawn_multiprocessing():
if requires_spawn_multiprocessing() or model_arch == "Glm5NextForCausalLM":
# The EngineCore subprocess re-imports the class and does not
# inherit the KV-cache patch above, so it OOMs. Run in-process
# so the patch applies.