biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[QeRL] Fix online quantized reloading (#38442)

Browse Source 
Signed-off-by: Kyle Sayers <kylesayrs@gmail.com>
This commit is contained in:
Kyle Sayers
2026-03-29 16:56:41 -04:00
committed by GitHub
parent 995dea1354
commit d28d86e8a3
9 changed files with 104 additions and 62 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
.buildkite/test_areas/models_distributed.yaml
View File

@@ -14,7 +14,7 @@ steps:
- tests/models/
commands:
- TARGET_TEST_SUITE=L4 pytest basic_correctness/ -v -s -m 'distributed(num_gpus=2)'
- CUDA_VISIBLE_DEVICES=0,1 pytest -v -s model_executor/model_loader/test_sharded_state_loader.py
- CUDA_VISIBLE_DEVICES=0,1 pytest -v -s model_executor/model_loader/test_sharded_state_loader.py -m '(not slow_test)'
# Avoid importing model tests that cause CUDA reinitialization error
- pytest models/test_transformers.py -v -s -m 'distributed(num_gpus=2)'
- pytest models/language -v -s -m 'distributed(num_gpus=2)'