[Model] add GLM-5.3-Flash support (#53906)
This commit is contained in:
@@ -41,6 +41,9 @@ from vllm.model_executor.warmup.qwen_triton_warmup import qwen_triton_warmup
|
||||
from vllm.model_executor.warmup.replayssm_warmup import (
|
||||
replayssm_autotune_warmup,
|
||||
)
|
||||
from vllm.model_executor.warmup.spec_decode_rejection_warmup import (
|
||||
spec_decode_rejection_warmup,
|
||||
)
|
||||
from vllm.platforms import current_platform
|
||||
from vllm.utils.deep_gemm import is_deep_gemm_supported
|
||||
from vllm.utils.flashinfer import has_flashinfer
|
||||
@@ -163,6 +166,7 @@ def kernel_warmup(worker: "Worker", *, process_local_only: bool = False):
|
||||
if worker.vllm_config.kernel_config.enable_jit_warmup:
|
||||
kimi_k3_triton_warmup(worker)
|
||||
fa4_cutedsl_warmup(worker)
|
||||
spec_decode_rejection_warmup(worker)
|
||||
qwen4_exp_qsa_triton_warmup(worker)
|
||||
|
||||
if current_platform.has_device_capability(90):
|
||||
|
||||
Reference in New Issue
Block a user