Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in
Toggle navigation
Menu
Open sidebar
OpenDAS
vllm_cscc
Commits
e99ba957
Unverified
Commit
e99ba957
authored
Feb 18, 2026
by
Aaron Hao
Committed by
GitHub
Feb 18, 2026
Browse files
[BUG] Fixing Weight Sync unit test (#34841)
Signed-off-by:
ahao-anyscale
<
ahao@anyscale.com
>
parent
64ac1395
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
6 additions
and
0 deletions
+6
-0
tests/entrypoints/weight_transfer/test_weight_transfer_llm.py
...s/entrypoints/weight_transfer/test_weight_transfer_llm.py
+6
-0
No files found.
tests/entrypoints/weight_transfer/test_weight_transfer_llm.py
View file @
e99ba957
...
@@ -124,6 +124,8 @@ def test_init_weight_transfer_engine_calls_engine():
...
@@ -124,6 +124,8 @@ def test_init_weight_transfer_engine_calls_engine():
if
torch
.
cuda
.
device_count
()
<
1
:
if
torch
.
cuda
.
device_count
()
<
1
:
pytest
.
skip
(
"Need at least 1 GPU for this test"
)
pytest
.
skip
(
"Need at least 1 GPU for this test"
)
# Run in-process so mock.patch works (spawn won't inherit the mock)
os
.
environ
[
"VLLM_ENABLE_V1_MULTIPROCESSING"
]
=
"0"
# Enable insecure serialization to allow pickling functions for collective_rpc
# Enable insecure serialization to allow pickling functions for collective_rpc
os
.
environ
[
"VLLM_ALLOW_INSECURE_SERIALIZATION"
]
=
"1"
os
.
environ
[
"VLLM_ALLOW_INSECURE_SERIALIZATION"
]
=
"1"
...
@@ -171,6 +173,8 @@ def test_update_weights_calls_engine():
...
@@ -171,6 +173,8 @@ def test_update_weights_calls_engine():
if
torch
.
cuda
.
device_count
()
<
1
:
if
torch
.
cuda
.
device_count
()
<
1
:
pytest
.
skip
(
"Need at least 1 GPU for this test"
)
pytest
.
skip
(
"Need at least 1 GPU for this test"
)
# Run in-process so mock.patch works (spawn won't inherit the mock)
os
.
environ
[
"VLLM_ENABLE_V1_MULTIPROCESSING"
]
=
"0"
# Enable insecure serialization to allow pickling functions for collective_rpc
# Enable insecure serialization to allow pickling functions for collective_rpc
os
.
environ
[
"VLLM_ALLOW_INSECURE_SERIALIZATION"
]
=
"1"
os
.
environ
[
"VLLM_ALLOW_INSECURE_SERIALIZATION"
]
=
"1"
...
@@ -228,6 +232,8 @@ def test_full_weight_transfer_flow():
...
@@ -228,6 +232,8 @@ def test_full_weight_transfer_flow():
if
torch
.
cuda
.
device_count
()
<
1
:
if
torch
.
cuda
.
device_count
()
<
1
:
pytest
.
skip
(
"Need at least 1 GPU for this test"
)
pytest
.
skip
(
"Need at least 1 GPU for this test"
)
# Run in-process so mock.patch works (spawn won't inherit the mock)
os
.
environ
[
"VLLM_ENABLE_V1_MULTIPROCESSING"
]
=
"0"
# Enable insecure serialization to allow pickling functions for collective_rpc
# Enable insecure serialization to allow pickling functions for collective_rpc
os
.
environ
[
"VLLM_ALLOW_INSECURE_SERIALIZATION"
]
=
"1"
os
.
environ
[
"VLLM_ALLOW_INSECURE_SERIALIZATION"
]
=
"1"
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment