-
Notifications
You must be signed in to change notification settings - Fork 695
Pull requests: vllm-project/vllm-ascend
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Add support for eagle layers with separate embed_tokens.
#5575
opened Jan 4, 2026 by
LumosLovegood
Loading…
[P/D]Remove mooncake kvpool unused parameter
local_hostname
#5574
opened Jan 4, 2026 by
LCAIZJ
Loading…
[Main2Main] Upgrade vllm commit to 0102
ci/build
documentation
Improvements or additions to documentation
module:tests
ready
read for review
ready-for-test
start test by label for PR
#5573
opened Jan 4, 2026 by
wjunLu
Loading…
[Graph][Fusion] Add AddRMSNormSPPattern and AddRMSNormSPPatternWithBias
#5569
opened Jan 4, 2026 by
ForBetterCodeNine
Loading…
[Refactor] Cleanup platform
module:core
module:tests
#5566
opened Jan 4, 2026 by
wangxiyuan
Loading…
[Bugfix] raise runtime error when npumodelrunner init failed
#5565
opened Jan 4, 2026 by
zhenwenqi2024
Loading…
[Bugfix] fix pcp + eplb error
module:core
module:tests
ready
read for review
ready-for-test
start test by label for PR
#5561
opened Dec 31, 2025 by
weiguihua2
Loading…
[Doc] add PaddleOCR-VL tutorials guide
documentation
Improvements or additions to documentation
#5556
opened Dec 31, 2025 by
zyz111222
Loading…
[Refactor] Modify the binding logic to allocate CPU cores for each NPU card
module:core
#5555
opened Dec 31, 2025 by
Rozwel-dx
Loading…
[Doc]modify the quantization user guide and add a quantization adaptation developer guide
documentation
Improvements or additions to documentation
#5554
opened Dec 31, 2025 by
InSec
Loading…
[Bugfix] Fix the graph capture failure issue in the eagle3+full scenario.
#5553
opened Dec 31, 2025 by
WithHades
Loading…
[kernel]EPLB:Adapt DispatchGmmCombineDecode operator to eplb tensor list and expert token numbers
module:core
module:ops
module:quantization
module:tests
#5552
opened Dec 31, 2025 by
wangyibo1005
Loading…
[WIP][Feature] Support MXFP8
module:core
module:ops
module:quantization
#5550
opened Dec 31, 2025 by
SlightwindSec
•
Draft
[Feat] enable hierarchical mc2 ops on A2 by default
module:core
module:ops
ready
read for review
ready-for-test
start test by label for PR
#5545
opened Dec 31, 2025 by
hwhaokun
Loading…
[Feature] add the magicmtp speculative decoding acceleration algorithm
module:ops
module:tests
#5542
opened Dec 31, 2025 by
chenaoxuan
Loading…
[P/D] Performance enhancement of Layerwise connector in TP asymmetric scenarios
#5540
opened Dec 30, 2025 by
liziyu179
Loading…
[EPLB]Eplb Config Renaming
documentation
Improvements or additions to documentation
merge-conflicts
module:core
module:tests
#5533
opened Dec 30, 2025 by
shenchuxiaofugui
Loading…
[main][refactor] Refactored the logic of update_attn_params
merge-conflicts
module:tests
#5532
opened Dec 30, 2025 by
drslark
Loading…
[feature]Token-Level Re-Inference for Fault Tolerance in vLLM-Ascend
#5530
opened Dec 30, 2025 by
Peter-Lu-22
Loading…
[Bugfix] Revert pr4214 multi-stream collect expert hotpot
merge-conflicts
module:ops
#5529
opened Dec 30, 2025 by
shenchuxiaofugui
Loading…
Previous Next
ProTip!
What’s not been updated in a month: updated:<2025-12-03.