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

[CI] Reorganize stage-b 1-GPU tests for 5090 compatibility (#16826)

Browse Source
This commit is contained in:
Alison Shao
2026-01-15 15:23:35 -08:00
committed by GitHub
parent 8b22deef5b
commit 146b5fcc84
136 changed files with 277 additions and 404 deletions
Download Patch File Download Diff File Expand all files Collapse all files
test/registered/mla/test_flashmla.py
+1 -1
View File
@@ -21,7 +21,7 @@ from sglang.test.test_utils import (
)
# FlashMLA attention backend tests with MTP speculative decoding
register_cuda_ci(est_time=284, suite="stage-b-test-small-1-gpu")
register_cuda_ci(est_time=284, suite="stage-b-test-large-1-gpu")
class TestFlashMLAAttnBackend(unittest.TestCase):
test/registered/mla/test_mla.py
+1 -1
View File
@@ -13,7 +13,7 @@ from sglang.test.test_utils import (
)
# MLA attention test with MGSM evaluation
register_cuda_ci(est_time=194, suite="stage-b-test-small-1-gpu")
register_cuda_ci(est_time=194, suite="stage-b-test-large-1-gpu")
register_amd_ci(est_time=1100, suite="stage-b-test-small-1-gpu-amd")
test/registered/mla/test_mla_deepseek_v3.py
+1 -1
View File
@@ -16,7 +16,7 @@ from sglang.test.test_utils import (
)
# DeepSeek-V3 MLA tests with torch compile, FA3, and MTP speculative decoding
register_cuda_ci(est_time=442, suite="stage-b-test-small-1-gpu")
register_cuda_ci(est_time=442, suite="stage-b-test-large-1-gpu")
register_amd_ci(
est_time=221,
suite="stage-b-test-small-1-gpu-amd",
test/registered/mla/test_mla_flashinfer.py
+1 -1
View File
@@ -15,7 +15,7 @@ from sglang.test.test_utils import (
)
# FlashInfer MLA backend tests with MTP speculative decoding
register_cuda_ci(est_time=302, suite="stage-b-test-small-1-gpu")
register_cuda_ci(est_time=302, suite="stage-b-test-large-1-gpu")
class TestFlashinferMLA(CustomTestCase):
test/registered/mla/test_mla_fp8.py
+1 -1
View File
@@ -13,7 +13,7 @@ from sglang.test.test_utils import (
)
# MLA FP8 KV cache test with MGSM evaluation
register_cuda_ci(est_time=77, suite="stage-b-test-small-1-gpu")
register_cuda_ci(est_time=77, suite="stage-b-test-large-1-gpu")
register_amd_ci(est_time=360, suite="stage-b-test-small-1-gpu-amd")
test/registered/mla/test_mla_int8_deepseek_v3.py
+1 -1
View File
@@ -16,7 +16,7 @@ from sglang.test.test_utils import (
)
# DeepSeek-V3 INT8 quantization tests (channel and block INT8)
register_cuda_ci(est_time=341, suite="stage-b-test-small-1-gpu")
register_cuda_ci(est_time=341, suite="stage-b-test-large-1-gpu")
class TestMLADeepseekV3ChannelInt8(CustomTestCase):