Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
Menu
Open sidebar
norm
vllm
Commits
c1c0d00b
Unverified
Commit
c1c0d00b
authored
Feb 27, 2024
by
Roy
Committed by
GitHub
Feb 26, 2024
Browse files
Don't use cupy when `enforce_eager=True` (#3037)
parent
d9f726c4
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
4 additions
and
1 deletion
+4
-1
vllm/engine/llm_engine.py
vllm/engine/llm_engine.py
+4
-1
No files found.
vllm/engine/llm_engine.py
View file @
c1c0d00b
...
@@ -284,7 +284,10 @@ class LLMEngine:
...
@@ -284,7 +284,10 @@ class LLMEngine:
is_driver_worker
=
True
,
is_driver_worker
=
True
,
)
)
self
.
_run_workers
(
"init_model"
,
cupy_port
=
get_open_port
())
# don't use cupy for eager mode
self
.
_run_workers
(
"init_model"
,
cupy_port
=
get_open_port
()
if
not
model_config
.
enforce_eager
else
None
)
self
.
_run_workers
(
self
.
_run_workers
(
"load_model"
,
"load_model"
,
max_concurrent_workers
=
self
.
parallel_config
.
max_concurrent_workers
=
self
.
parallel_config
.
...
...
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