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 15, 11:00 UTC
cleanup invalid prints (#18050)
Signed-off-by: calvin chen <120380290@qq.com>
Calvin Chen committed
May 13, 2025 at 14:01 UTC
48545728d872e9311b4a87b7976673fc7d1fb554
1 file changed
-2
vllm/worker/hpu_model_runner.py
-2
@@ -1554,10 +1554,8 @@ class HPUModelRunnerBase(ModelRunnerBase[TModelInputForHPU]):
1554
'Please update Gaudi Software Suite.')
1555
with compile_only_mode_context(
1556
) if can_use_compile_only_mode else contextlib.nullcontext():
1557
- print("aa")
1557
self.warmup_all_buckets(self.bucketing_ctx.prompt_buckets, True,
1558
kv_caches)
1560
- print("bb")
1559
self.warmup_all_buckets(self.bucketing_ctx.decode_buckets, False,
1560
kv_caches)
1561