minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 13 Packages Projects Releases Wiki Activity

ci: B200 conditional split + LPT_SLOP removal (stage-c partition 8→3) (#25203)

Browse Source
This commit is contained in:
Alison Shao
2026-05-13 18:40:45 -07:00
committed by GitHub
parent 5fb6bde6c0
commit 1e308aec66
19 changed files with 33 additions and 34 deletions
Download Patch File Download Diff File Expand all files Collapse all files
test/registered/quant/test_deepseek_v32_fp4_mtp_4gpu.py
+1 -1
View File
@@ -16,7 +16,7 @@ from sglang.test.test_utils import (
)
register_cuda_ci(
est_time=1060,
est_time=690,
stage="stage-c",
runner_config="4-gpu-b200",
)
test/registered/quant/test_deepseek_v3_fp4_4gpu.py
+1 -1
View File
@@ -14,7 +14,7 @@ from sglang.test.test_utils import (
write_github_step_summary,
)
register_cuda_ci(est_time=1190, stage="stage-c", runner_config="4-gpu-b200")
register_cuda_ci(est_time=960, stage="stage-c", runner_config="4-gpu-b200")
FULL_DEEPSEEK_V3_FP4_MODEL_PATH = "nvidia/DeepSeek-V3-0324-FP4"
SERVER_LAUNCH_TIMEOUT = 1200
test/registered/quant/test_fp8_blockwise_gemm.py
+1 -1
View File
@@ -12,7 +12,7 @@ from sglang.test.test_utils import (
try_cached_model,
)
register_cuda_ci(est_time=630, stage="stage-c", runner_config="4-gpu-b200")
register_cuda_ci(est_time=430, suite="nightly-4-gpu-b200", nightly=True)
MODEL_PATH = "Qwen/Qwen3-4B-Instruct-2507-FP8"
MXFP8_MODEL_PATH = "zianglih/Qwen3-4B-Instruct-2507-MXFP8"
test/registered/quant/test_nvfp4_gemm.py
+1 -1
View File
@@ -12,7 +12,7 @@ from sglang.test.test_utils import (
try_cached_model,
)
register_cuda_ci(est_time=420, stage="stage-c", runner_config="4-gpu-b200")
register_cuda_ci(est_time=350, stage="stage-c", runner_config="4-gpu-b200")
MODEL_PATH = "nvidia/Llama-3.1-8B-Instruct-NVFP4"