yangjianfengo1 8e1b35a09b 【Fix bug] w4afp8 的nblock固定为256，并且fa3的append attn 增加mask参数 (#3771) ...

* fix w4afp8

* 增加集中式配置

* codestyle

* fix fa3 append attn

2025-09-02 19:17:01 +08:00

..

cache_manager

…

demo

…

distributed

…

engine

[BugFix] fix error of import paddle.base.core.Config (#3761)

2025-09-02 17:23:27 +08:00

entrypoints

[Feature] mm and thinking model support structred output (#2749)

2025-09-02 16:21:09 +08:00

input

…

inter_communicator

…

logger

…

metrics

…

model_executor

【Fix bug] w4afp8 的nblock固定为256，并且fa3的append attn 增加mask参数 (#3771)

2025-09-02 19:17:01 +08:00

multimodal

…

output

…

platforms

…

plugins

…

reasoning

…

rl

[Feature] support model weight update in ep (#3765)

2025-09-02 17:16:03 +08:00

scheduler

…

spec_decode

…

splitwise

…

worker

[Feature] support model weight update in ep (#3765)

2025-09-02 17:16:03 +08:00

__init__.py

…

config.py

[Feature] support model weight update in ep (#3765)

2025-09-02 17:16:03 +08:00

envs.py

…

import_ops.py

…

stop.sh

…

test.yaml

…

utils.py

…