lzy d339df2e90 Supports DP+TP+EP hybrid parallel deployment strategy (#3489) ...

* Support DP+TP+EP hybrid parallel deployment strategy

* Support DP+TP+EP hybrid parallel deployment strategy

* fix conflict

* add moe_tp_ep function split_allgather_out

* del tp_group in moe_cutlass_backend

* for ci

* fix parallel_config for ci

* del log

2025-08-26 00:04:01 -07:00

..

cache_manager

add error traceback info (#3419)

2025-08-19 19:32:04 +08:00

demo

polish code with new pre-commit rule (#2923)

2025-07-19 23:19:27 +08:00

distributed

Supports DP+TP+EP hybrid parallel deployment strategy (#3489)

2025-08-26 00:04:01 -07:00

engine

[Feature] bad words support v1 scheduler and specifiy token ids (#3608)

2025-08-25 20:14:51 -07:00

entrypoints

[Feature] bad words support v1 scheduler and specifiy token ids (#3608)

2025-08-25 20:14:51 -07:00

input

[Feature] bad words support v1 scheduler and specifiy token ids (#3608)

2025-08-25 20:14:51 -07:00

inter_communicator

add error traceback info (#3419)

2025-08-19 19:32:04 +08:00

logger

add error log to file (#3431)

2025-08-20 09:52:34 +08:00

metrics

[Trace]add trace when fd start (#3174)

2025-08-05 21:18:27 +08:00

model_executor

Supports DP+TP+EP hybrid parallel deployment strategy (#3489)

2025-08-26 00:04:01 -07:00

multimodal

Unify server-side and model-side Config(Part-5) (#3497)

2025-08-21 19:00:21 +08:00

output

[Feature][MTP]support new speculative decoding method named hybrid mtp with ngram (#3610)

2025-08-26 14:29:22 +08:00

platforms

add error traceback info (#3419)

2025-08-19 19:32:04 +08:00

plugins

[plugin] Custom model_runner/model support (#3186)

2025-08-04 18:52:39 -07:00

reasoning

[Feature] add tool parser (#3483)

2025-08-21 17:25:44 +08:00

rl

[V1 Loader] support weight_only (#3413)

2025-08-23 13:13:41 +08:00

scheduler

add error traceback info (#3419)

2025-08-19 19:32:04 +08:00

spec_decode

[Feature][MTP]support new speculative decoding method named hybrid mtp with ngram (#3610)

2025-08-26 14:29:22 +08:00

splitwise

Unify server-side and model-side Config(Part-5) (#3497)

2025-08-21 19:00:21 +08:00

worker

Supports DP+TP+EP hybrid parallel deployment strategy (#3489)

2025-08-26 00:04:01 -07:00

__init__.py

[BugFix] Fix default log level of paddleformers (#3376)

2025-08-14 11:36:24 +08:00

config.py

Supports DP+TP+EP hybrid parallel deployment strategy (#3489)

2025-08-26 00:04:01 -07:00

envs.py

[Features] support hugging face qwen3 dense and qwen2 model (#3574)

2025-08-26 10:54:53 +08:00

import_ops.py

fix import error (#2944)

2025-07-22 14:06:01 +08:00

stop.sh

polish code with new pre-commit rule (#2923)

2025-07-19 23:19:27 +08:00

test.yaml

[Sync] Update to latest code (#2679)

2025-07-03 15:43:53 +08:00

utils.py

[Executor] CUDAGraph support RL training (#3265)

2025-08-25 20:59:30 +08:00