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

[misc] Use --cuda-graph-max-bs-decode in tests, examples, and docs (#29591)

Browse Source
This commit is contained in:
Liangsheng Yin
2026-06-28 18:38:28 -07:00
committed by GitHub
parent 3217410cf6
commit 909123ddb8
140 changed files with 304 additions and 290 deletions
Download Patch File Download Diff File Expand all files Collapse all files
test/registered/mla/test_flashmla.py
+1 -1
View File
@@ -32,7 +32,7 @@ class TestFlashMLAMTP(CustomTestCase):
if torch.cuda.is_available() and torch.version.cuda:
other_args.extend(
[
"--cuda-graph-max-bs",
"--cuda-graph-max-bs-decode",
"4",
"--disable-radix",
"--enable-torch-compile",
test/registered/mla/test_mla_flashinfer.py
+1 -1
View File
@@ -27,7 +27,7 @@ class TestFlashinferMLAMTP(CustomTestCase):
if torch.cuda.is_available() and torch.version.cuda:
other_args.extend(
[
"--cuda-graph-max-bs",
"--cuda-graph-max-bs-decode",
"4",
"--enable-torch-compile",
"--torch-compile-max-bs",
test/registered/mla/test_mla_int8_deepseek_v3.py
+2 -2
View File
@@ -28,7 +28,7 @@ class TestDeepseekV3MTPChannelInt8(CustomTestCase):
if torch.cuda.is_available() and torch.version.cuda:
other_args.extend(
[
"--cuda-graph-max-bs",
"--cuda-graph-max-bs-decode",
"16",
"--enable-torch-compile",
"--torch-compile-max-bs",
@@ -91,7 +91,7 @@ class TestDeepseekV3MTPBlockInt8(CustomTestCase):
if torch.cuda.is_available() and torch.version.cuda:
other_args.extend(
[
"--cuda-graph-max-bs",
"--cuda-graph-max-bs-decode",
"16",
"--enable-torch-compile",
"--torch-compile-max-bs",