A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 12, 14:45 UTC
Fix PoolingParams.skip_reading_prefix_cache type (#29364)
Signed-off-by: KFL <kludev@gmail.com>
kflu committed
Nov 24, 2025 at 22:39 UTC
ce58fdc1c366b0257c2b2d8310b14d4ea8f8dd30
2 files changed
+2
-2
vllm/pooling_params.py
+1
-1
@@ -57,7 +57,7 @@ class PoolingParams(
57
## Internal use only
58
task: PoolingTask | None = None
59
requires_token_ids: bool = False
60
- skip_reading_prefix_cache: bool = None
60
+ skip_reading_prefix_cache: bool | None = None
61
extra_kwargs: dict[str, Any] | None = None
62
output_kind: RequestOutputKind = RequestOutputKind.FINAL_ONLY
63
vllm/sampling_params.py
+1
-1
@@ -238,7 +238,7 @@ class SamplingParams(
238
generated token can complete the sequence."""
239
_bad_words_token_ids: list[list[int]] | None = None
240
241
- skip_reading_prefix_cache: bool = None
241
+ skip_reading_prefix_cache: bool | None = None
242
243
@staticmethod
244
def from_optional(