Skip to content

Commit df5d935

Browse files
tedzhouhknv-anants
andauthored
fix: add mem frac for sglang dsr1 8gpu (CP #5260) (#5282)
Signed-off-by: hongkuanz <hongkuanz@nvidia.com> Co-authored-by: Anant Sharma <anants@nvidia.com>
1 parent e11e5cd commit df5d935

File tree

1 file changed

+4
-0
lines changed

1 file changed

+4
-0
lines changed

recipes/deepseek-r1/sglang/disagg-8gpu/deploy.yaml

Lines changed: 4 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -60,6 +60,8 @@ spec:
6060
- decode
6161
- --disaggregation-bootstrap-port
6262
- "30001"
63+
- --mem-fraction-static
64+
- "0.75"
6365
- --host
6466
- 0.0.0.0
6567
- --prefill-round-robin-balance
@@ -97,6 +99,8 @@ spec:
9799
- prefill
98100
- --disaggregation-bootstrap-port
99101
- "30001"
102+
- --mem-fraction-static
103+
- "0.75"
100104
- --host
101105
- 0.0.0.0
102106
- --load-balance-method

0 commit comments

Comments
 (0)