[AscendScheduler][Bugfix] Remove num_draft_tokens while allocating slots #1718
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
wangxiyuan
reviewed
Jul 10, 2025
| @@ -282,15 +282,10 @@ def skip_cur_request(): | |||
| req_index += 1 | |||
| continue | |||
|
|
|||
| num_draft_tokens = max( | |||
There was a problem hiding this comment.
please use vllm_version_is to keep it both work on 0.9.2 and main
Signed-off-by: MengqingCao <cmq0113@163.com>
MengqingCao
force-pushed
the
ascendscheduler
branch
from
d5051ff to
06be93e
Compare
github-actions
bot
added
the
documentation
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #1718 +/- ##
===========================================
+ Coverage 27.39% 54.57% +27.18%
===========================================
Files 56 80 +24
Lines 6191 9968 +3777
===========================================
+ Hits 1696 5440 +3744
- Misses 4495 4528 +33
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
wangxiyuan
approved these changes
Jul 10, 2025
ApsarasX
approved these changes
Jul 10, 2025
Signed-off-by: MengqingCao <cmq0113@163.com>
wangxiyuan
reviewed
Jul 10, 2025
| MODELS = [ | ||
| "Qwen/Qwen2.5-0.5B-Instruct", | ||
| # TODO: REVERT ME when oom is fixed | ||
| # "vllm-ascend/Qwen3-30B-A3B-Puring" |
There was a problem hiding this comment.
Just tested locally, the test passed. I guess it's the resource release problem on CI system.
There was a problem hiding this comment.
Maybe, will raise another pr to fix it
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
What this PR does / why we need it?
Now there is no need to calculate
num_draft_tokenswhen allocating slots.This PR follows the changes in vllm: vllm-project/vllm#20701
Does this PR introduce any user-facing change?
N/A
How was this patch tested?
CI passed with existing test