xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-07-21 16:27:19 +08:00
Code Issues Packages Projects Releases Wiki Activity

[BUGFIX] When skip_tokenize_init and multistep are set, execution crashes (#12277)

Browse Source 
Signed-off-by: maleksan85 <maleksan@amd.com>
Co-authored-by: maleksan85 <maleksan@amd.com>
This commit is contained in:
Aleksandr Malyshev 2025-01-21 15:30:46 -08:00 committed by GitHub
parent 2acba47d9b
commit 69196a9bc7
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/engine/output_processor/multi_step.py
View File

@ -144,7 +144,7 @@ class MultiStepOutputProcessor(SequenceGroupOutputProcessor):
def _process_decode_and_stop(self, seq: Sequence, def _process_decode_and_stop(self, seq: Sequence,
sampling_params: SamplingParams) -> None: sampling_params: SamplingParams) -> None:
new_char_count = 0 new_char_count = 0
if sampling_params.detokenize: if sampling_params.detokenize and self.detokenizer:
new_char_count = self.detokenizer.decode_sequence_inplace( new_char_count = self.detokenizer.decode_sequence_inplace(
seq, sampling_params) seq, sampling_params)