EngineX-Hygon/sglang
5
0
Fork 0
Code Issues Pull Requests Actions 7 Projects Releases Wiki Activity

Fix oom issues with fp8 for llama (#1454)

Browse Source
This commit is contained in:
Lianmin Zheng
2024-09-18 03:45:19 -07:00
committed by GitHub
parent aa2750beb3
commit 1acccb364a
8 changed files with 33 additions and 21 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
python/sglang/test/test_utils.py
View File

@@ -22,6 +22,7 @@ from sglang.lang.backend.runtime_endpoint import RuntimeEndpoint
from sglang.srt.utils import kill_child_process
from sglang.utils import get_exception_traceback
DEFAULT_FP8_MODEL_NAME_FOR_TEST = "neuralmagic/Meta-Llama-3.1-8B-FP8"
DEFAULT_MODEL_NAME_FOR_TEST = "meta-llama/Meta-Llama-3.1-8B-Instruct"
DEFAULT_MOE_MODEL_NAME_FOR_TEST = "mistralai/Mixtral-8x7B-Instruct-v0.1"
DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH = 600