-
Notifications
You must be signed in to change notification settings - Fork 238
Pull requests: vllm-project/vllm-ascend
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
[Bugfix][CI] Remove V0 Spec Decode CI
long-term-test
enable long term test for PR
module:tests
ready-for-test
start test by label for PR
#1656
opened Jul 7, 2025 by
shen-shanshan
Loading…
[CI/Build] Upgrade CANN to 8.2.RC1.alpha003
documentation
Improvements or additions to documentation
#1653
opened Jul 7, 2025 by
MengqingCao
Loading…
Upgrade vLLM version to v0.9.2
accuracy-test
enable all accuracy test for PR
documentation
Improvements or additions to documentation
ready-for-test
start test by label for PR
[V0.9.1][BugFix] Fix load weight error and add new e2e case
module:tests
#1651
opened Jul 7, 2025 by
shikang-hangzhou
Loading…
[CustomOP][Refactor] Register CustomOP instead of overwrite forward_oot
module:ops
module:quantization
#1647
opened Jul 7, 2025 by
MengqingCao
•
Draft
[CI/UT][Refactor] Refactor multi-card CI
module:tests
#1645
opened Jul 7, 2025 by
MengqingCao
•
Draft
[0.9.1][PD][Perf] Avoid performing cpu all_reduce in disaggregated-prefill scenario.
#1644
opened Jul 7, 2025 by
whx-sjtu
Loading…
Enable the super kernel feature under the Multistream Moe feature
module:core
module:ops
module:quantization
#1641
opened Jul 7, 2025 by
NNUCJ
Loading…
[FOLLOWUP] Use base test to avoid patch everwhere
module:tests
ready
read for review
#1634
opened Jul 6, 2025 by
Yikun
Loading…
[BUGFIX] FIX mtp accuraccy when temperture is not 0
module:tests
#1632
opened Jul 5, 2025 by
JC-ut0
Loading…
[0.9.1][Perf] Optimize the number of rope-related index selections in deepseek.
#1614
opened Jul 3, 2025 by
whx-sjtu
Loading…
[CI][Benchmark] Add Qwen3-30B-A3B and Qwen3-32B performance benchmark
#1613
opened Jul 3, 2025 by
Potabk
Loading…
[0.9.1][Perf] Launch load kv task asynchronously with thread pool.
#1612
opened Jul 3, 2025 by
ganyi1996ppo
Loading…
[bugfix] fix graph_batch_sizes padding bug
merge-conflicts
#1607
opened Jul 3, 2025 by
zzzzwwjj
Loading…
[Refactor] Use tuple as kv cache instead of tensor
merge-conflicts
module:ops
#1594
opened Jul 2, 2025 by
lidenghui1110
Loading…
[Feature] Optimize forward metadata collection across dp ranks
performance-test
enable performance test for PR
ready
read for review
ready-for-test
start test by label for PR
#1593
opened Jul 2, 2025 by
jianzs
Loading…
Previous Next
ProTip!
Follow long discussions with comments:>50.