xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-15 18:16:46 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Bugfix][Misc]: fix graph capture for decoder (#9549)

Browse Source
This commit is contained in:
yudian0504 2024-10-22 01:33:30 +08:00 committed by GitHub
parent f6b97293aa
commit 8ca8954841
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/worker/model_runner.py
View File

@ -828,7 +828,7 @@ class ModelInputForGPUBuilder(ModelRunnerInputBuilderBase[ModelInputForGPU]):
cuda_graph_pad_size = self._get_cuda_graph_pad_size( cuda_graph_pad_size = self._get_cuda_graph_pad_size(
num_seqs=len(seq_lens), num_seqs=len(seq_lens),
max_decode_seq_len=max_encoder_seq_len, max_decode_seq_len=max_decode_seq_len,
max_encoder_seq_len=max_encoder_seq_len) max_encoder_seq_len=max_encoder_seq_len)
batch_size = len(input_tokens) batch_size = len(input_tokens)