xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-01-11 03:04:28 +08:00
Code Issues Packages Projects Releases Wiki Activity

Update sampling_metadata.py (#21937)

Browse Source 
Signed-off-by: Aviad Rossmann <aviadr@neureality.ai>
This commit is contained in:
Aviad Rossmann 2025-08-01 09:23:18 +03:00 committed by GitHub
parent 61dcc280fa
commit 53d7c39271
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 6 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
vllm/model_executor/sampling_metadata.py
View File

@ -539,37 +539,37 @@ class SamplingTensors:
temperatures_t = torch.tensor(
temperatures,
device="cpu",
dtype=dtype,
dtype=torch.float32,
pin_memory=pin_memory,
)
top_ps_t = torch.tensor(
top_ps,
device="cpu",
dtype=dtype,
dtype=torch.float32,
pin_memory=pin_memory,
)
min_ps_t = torch.tensor(
min_ps,
device="cpu",
dtype=dtype,
dtype=torch.float32,
pin_memory=pin_memory,
)
presence_penalties_t = torch.tensor(
presence_penalties,
device="cpu",
dtype=dtype,
dtype=torch.float32,
pin_memory=pin_memory,
)
frequency_penalties_t = torch.tensor(
frequency_penalties,
device="cpu",
dtype=dtype,
dtype=torch.float32,
pin_memory=pin_memory,
)
repetition_penalties_t = torch.tensor(
repetition_penalties,
device="cpu",
dtype=dtype,
dtype=torch.float32,
pin_memory=pin_memory,
)
top_ks_t = torch.tensor(