minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 13 Packages Projects Releases Wiki Activity

fix(mlx): default prefill_aware_swa=False on MlxModelRunnerStub (#29681)

Browse Source
This commit is contained in:
Tai An
2026-06-30 22:55:43 -07:00
committed by GitHub
parent 3cdc2415b1
commit 546d6e23ad
1 changed files with 8 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
python/sglang/srt/hardware_backend/mlx/model_runner_stub.py
+8
View File
@@ -85,6 +85,14 @@ class MlxModelRunnerStub(ModelRunner):
# AttributeError. # AttributeError.
canary_manager = None canary_manager = None
# No prefill-aware SWA on the MLX path. The base ModelRunner derives this in
# its full initialize() from `model.is_prefill_aware_swa()`, which this
# lightweight override skips (and `_DummyModel` does not implement). The
# scheduler reads `model_runner.prefill_aware_swa` unconditionally when
# admitting a prefill batch, so default to False as a class attribute to keep
# that path working instead of raising AttributeError.
prefill_aware_swa = False
def __init__(self, *args, mlx_pool_size: int | None = None, **kwargs): def __init__(self, *args, mlx_pool_size: int | None = None, **kwargs):
self._mlx_pool_size = mlx_pool_size self._mlx_pool_size = mlx_pool_size
super().__init__(*args, **kwargs) super().__init__(*args, **kwargs)