xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-01-24 00:14:34 +08:00
Code Issues Packages Projects Releases Wiki Activity

[TPU][Bugfix] fix OOM issue in CI test (#21550)

Browse Source 
Signed-off-by: Chengji Yao <chengjiyao@google.com>
This commit is contained in:
Chengji Yao 2025-07-24 23:01:53 -07:00 committed by GitHub
parent 85d051f026
commit 40d86ee412
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
tests/v1/tpu/test_basic.py
View File

@ -59,7 +59,7 @@ def test_basic(
# actually test chunked prompt
max_num_batched_tokens=1024,
max_model_len=8192,
gpu_memory_utilization=0.7,
gpu_memory_utilization=0.95,
max_num_seqs=max_num_seqs,
tensor_parallel_size=tensor_parallel_size) as vllm_model:
vllm_outputs = vllm_model.generate_greedy(example_prompts,