minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

[CI] slash handler: lookup runs_on from runner_configs.yml (#25394)

Browse Source
This commit is contained in:
Liangsheng Yin
2026-05-15 13:59:00 -07:00
committed by GitHub
parent 54221dd998
commit f9caf43095
33 changed files with 261 additions and 309 deletions
Download Patch File Download Diff File Expand all files Collapse all files
test/manual/quant/test_autoround.py
-3
View File
@@ -7,7 +7,6 @@ import unittest
from types import SimpleNamespace
from sglang.srt.utils import kill_process_tree
from sglang.test.ci.ci_register import register_cuda_ci
from sglang.test.run_eval import run_eval
from sglang.test.test_utils import (
DEFAULT_AUTOROUND_MODEL_NAME_FOR_TEST,
@@ -17,8 +16,6 @@ from sglang.test.test_utils import (
popen_launch_server,
)
register_cuda_ci(est_time=99, suite="stage-b-test-1-gpu-large")
class TestAutoRound(CustomTestCase):
@classmethod
test/manual/quant/test_deepseek_v32_fp4_4gpu.py
-3
View File
@@ -2,7 +2,6 @@ import unittest
from types import SimpleNamespace
from sglang.srt.utils import kill_process_tree
from sglang.test.ci.ci_register import register_cuda_ci
from sglang.test.run_eval import run_eval
from sglang.test.send_one import BenchArgs, send_one_prompt
from sglang.test.test_utils import (
@@ -13,8 +12,6 @@ from sglang.test.test_utils import (
write_github_step_summary,
)
register_cuda_ci(est_time=874, suite="stage-c-test-4-gpu-b200")
FULL_DEEPSEEK_V3_FP4_MODEL_PATH = "nvidia/DeepSeek-V3.2-NVFP4"
SERVER_LAUNCH_TIMEOUT = 1200
test/manual/quant/test_eval_fp8_accuracy.py
-4
View File
@@ -2,7 +2,6 @@ import unittest
from types import SimpleNamespace
from sglang.srt.utils import is_hip, kill_process_tree
from sglang.test.ci.ci_register import register_amd_ci, register_cuda_ci
from sglang.test.run_eval import run_eval
from sglang.test.test_utils import (
DEFAULT_MODEL_NAME_FOR_ACCURACY_TEST_FP8,
@@ -14,9 +13,6 @@ from sglang.test.test_utils import (
popen_launch_server,
)
register_cuda_ci(est_time=351, suite="stage-b-test-1-gpu-large")
register_amd_ci(est_time=600, suite="stage-b-test-1-gpu-small-amd")
class TestEvalFP8Accuracy(CustomTestCase):
@classmethod
test/manual/quant/test_quantization.py
-3
View File
@@ -4,7 +4,6 @@ import warnings
from types import SimpleNamespace
from sglang.srt.utils import kill_process_tree
from sglang.test.ci.ci_register import register_cuda_ci
from sglang.test.run_eval import run_eval
from sglang.test.test_utils import (
DEFAULT_MODEL_NAME_FOR_NIGHTLY_EVAL_QUANT_TP1,
@@ -16,8 +15,6 @@ from sglang.test.test_utils import (
write_results_to_json,
)
register_cuda_ci(est_time=460, suite="stage-b-test-1-gpu-large")
MODEL_SCORE_THRESHOLDS = {
# Baselines observed with gsm8k 5-shot concatenated format via chat API,
# which scores lower than reported benchmarks using proper CoT format.