chenchenghao/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 11 Packages Projects Releases Wiki Activity

Merge stage-c-test-large-4-gpu suites into partitioned suites (#18325)

Browse Source
This commit is contained in:
Alison Shao
2026-02-06 15:32:33 -08:00
committed by GitHub
parent f1ff697494
commit bedade1ef0
10 changed files with 8 additions and 97 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
test/registered/attention/test_local_attn.py
View File

@@ -16,7 +16,7 @@ from sglang.test.test_utils import (
)
# Local attention with FA3 (requires SM 90+ / H100, tp=4)
register_cuda_ci(est_time=200, suite="stage-c-test-large-4-gpu")
register_cuda_ci(est_time=200, suite="stage-c-test-4-gpu-h100")
@unittest.skipIf(get_device_sm() < 90, "Test requires CUDA SM 90 or higher")

2
test/registered/distributed/test_dp_attention_large.py
View File

@@ -23,7 +23,7 @@ from sglang.test.test_utils import (
popen_launch_server,
)
register_cuda_ci(est_time=350, suite="stage-c-test-large-4-gpu")
register_cuda_ci(est_time=350, suite="stage-c-test-4-gpu-h100")
class TestDPAttentionDP2TP4(

2
test/registered/moe/test_cutedsl_moe.py
View File

@@ -12,7 +12,7 @@ from sglang.srt.layers.moe.flashinfer_cutedsl_moe import flashinfer_cutedsl_moe_
from sglang.srt.layers.moe.topk import TopKConfig, select_experts
from sglang.test.ci.ci_register import register_cuda_ci
register_cuda_ci(est_time=300, suite="stage-c-test-large-4-gpu-b200")
register_cuda_ci(est_time=300, suite="stage-c-test-4-gpu-b200")
SKIP_TEST = torch.cuda.get_device_capability() < (10, 0)
SKIP_REASON = "Nvfp4 Requires compute capability of 10 or above."

2
test/registered/rl/test_return_routed_experts.py
View File

@@ -21,7 +21,7 @@ from sglang.test.test_utils import (
popen_launch_server,
)
register_cuda_ci(est_time=360, suite="stage-c-test-large-4-gpu")
register_cuda_ci(est_time=360, suite="stage-c-test-4-gpu-h100")
SHAREGPT_URL = (
"https://huggingface.co/datasets/anon8231489123/"

2
test/registered/spec/eagle/test_eagle_dp_attention.py
View File

@@ -20,7 +20,7 @@ from sglang.test.test_utils import (
)
# EAGLE3 with DP attention (tp=2, dp=2, requires 4 GPUs)
register_cuda_ci(est_time=200, suite="stage-c-test-large-4-gpu")
register_cuda_ci(est_time=200, suite="stage-c-test-4-gpu-h100")
class TestEAGLE3EngineDPAttention(CustomTestCase):

2
test/registered/spec/eagle/test_eagle_infer_beta_dp_attention.py
View File

@@ -17,7 +17,7 @@ from sglang.test.test_utils import (
)
# EAGLE with DP attention on B200 (tp=2, dp=2, requires 4 B200 GPUs)
register_cuda_ci(est_time=300, suite="stage-c-test-large-4-gpu-b200")
register_cuda_ci(est_time=300, suite="stage-c-test-4-gpu-b200")
def test_gsm8k(base_url: str):