Fix 8-gpu B200 nightly tests (#13457)
This commit is contained in:
@@ -25,8 +25,8 @@ class TestNightlyDeepseekV31Performance(unittest.TestCase):
|
||||
"--trust-remote-code",
|
||||
"--tp",
|
||||
"8",
|
||||
"--dp",
|
||||
"8",
|
||||
"--model-loader-extra-config",
|
||||
'{"enable_multithread_load": true}',
|
||||
],
|
||||
},
|
||||
{
|
||||
@@ -35,8 +35,6 @@ class TestNightlyDeepseekV31Performance(unittest.TestCase):
|
||||
"--trust-remote-code",
|
||||
"--tp",
|
||||
"8",
|
||||
"--dp",
|
||||
"8",
|
||||
"--speculative-algorithm",
|
||||
"EAGLE",
|
||||
"--speculative-num-steps",
|
||||
@@ -47,6 +45,8 @@ class TestNightlyDeepseekV31Performance(unittest.TestCase):
|
||||
"4",
|
||||
"--mem-frac",
|
||||
"0.7",
|
||||
"--model-loader-extra-config",
|
||||
'{"enable_multithread_load": true}',
|
||||
],
|
||||
},
|
||||
]
|
||||
|
||||
@@ -25,8 +25,8 @@ class TestNightlyDeepseekV32Performance(unittest.TestCase):
|
||||
"--trust-remote-code",
|
||||
"--tp",
|
||||
"8",
|
||||
"--dp",
|
||||
"8",
|
||||
"--model-loader-extra-config",
|
||||
'{"enable_multithread_load": true}',
|
||||
],
|
||||
},
|
||||
{
|
||||
@@ -35,8 +35,6 @@ class TestNightlyDeepseekV32Performance(unittest.TestCase):
|
||||
"--trust-remote-code",
|
||||
"--tp",
|
||||
"8",
|
||||
"--dp",
|
||||
"8",
|
||||
"--speculative-algorithm",
|
||||
"EAGLE",
|
||||
"--speculative-num-steps",
|
||||
@@ -47,6 +45,8 @@ class TestNightlyDeepseekV32Performance(unittest.TestCase):
|
||||
"4",
|
||||
"--mem-frac",
|
||||
"0.7",
|
||||
"--model-loader-extra-config",
|
||||
'{"enable_multithread_load": true}',
|
||||
],
|
||||
},
|
||||
{
|
||||
@@ -55,14 +55,14 @@ class TestNightlyDeepseekV32Performance(unittest.TestCase):
|
||||
"--trust-remote-code",
|
||||
"--tp",
|
||||
"8",
|
||||
"--dp",
|
||||
"8",
|
||||
"--attention-backend",
|
||||
"nsa",
|
||||
"--nsa-prefill-backend",
|
||||
"flashmla_sparse",
|
||||
"--nsa-decode-backend",
|
||||
"flashmla_kv",
|
||||
"--model-loader-extra-config",
|
||||
'{"enable_multithread_load": true}',
|
||||
],
|
||||
},
|
||||
]
|
||||
|
||||
Reference in New Issue
Block a user