COMMITS
August 29, 2026
V
[Doc][Model] Add Hy4-preview deployment guide (#15281)
Void Velocity committed
T
[TEST]Migrate Qwen3 and DeepSeek test cases from weekly to nightly (#15282)
taozi01319 committed
Z
[CI] Fix shallow vLLM range recovery (#15306)
zhao-stack committed
Z
J
[Test] Fix UT for EPLBController (#15283)
JIACHENG XU committed
A
C
[Test][Misc] Move 2 size cases to double_node (#15141)
chen-commits committed
Z
[bugfix] roll back megamoe op (#15267)
zzzzwwjj committed
Q
[Performance](attention): fuse SFA DCP output all-to-all (#14457)
Qiu Chunshuo committed
Z
[CI] Check vLLM PR compatibility with vllm-ascend (#14560)
zhao-stack committed
Z
[CI] upgrade a5 wheel manylinux (#15260)
zhangxinyuehfad committed
T
[TEST]Modification of Test Cases for minimax27,qwen3,deepseek (#15019)
taozi01319 committed
Z
[Misc] remove compatibility with dispatch_v1/combine_v1 interface (#15157)
zzzzwwjj committed
J
[Feature][EPLB] Support EPLB for MiniMax-M3 (#14973)
JIACHENG XU committed
J
[Test] Supplement UT for EPLB (#14929)
JIACHENG XU committed
W
[Misc] Update Ascendconfig docstring (#15058)
wangxiyuan committed
J
[BugFix] Release unused unquantized weights for fused MC2 (#15216)
Jade Zheng committed
K
[Refactor][Config] Remove MLAPO environment variable (#14829)
kk_1999 committed
H
[Feature][ModelRunner V1] Add score encoder cache manager (#7500)
hotTea committed
W
S
[Refactor][Device][3/N] Migrate core runtime selection to hardware profiles (#14076)
Shaoxu Cheng committed
August 28, 2026
V
[Revert] Revert "[CI] Add .dockerignore and increase pip --timeout to 1200s" (#14964) (#15236)
vllm-ascend-ci committed
F
W
Q
G
G
J