PaddlePaddle/PaddleFleet v0.1.0

PaddleFleet version 0.1.0

Repository: PaddlePaddle/PaddleFleet

Tag: v0.1.0

Published: 2026-01-18T11:08:30Z

Prerelease: no

Release notes: The first version for PaddleFleet.

What's Changed

• Add preliminary code by @From00 in https://github.com/PaddlePaddle/PaddleFleet/pull/1
• add spec utils by @FeixLiu in https://github.com/PaddlePaddle/PaddleFleet/pull/2
• add enums.py and identity_op.py by @GuoxiaWang in https://github.com/PaddlePaddle/PaddleFleet/pull/3
• add vpp_simulator by @Waynezee in https://github.com/PaddlePaddle/PaddleFleet/pull/5
• PaddleFleet distributed initialization and ProcessGroup create. by @Hz188 in https://github.com/PaddlePaddle/PaddleFleet/pull/8
• add codestyle workflow by @swgu98 in https://github.com/PaddlePaddle/PaddleFleet/pull/6
• Complete parallel_state.py by @Hz188 in https://github.com/PaddlePaddle/PaddleFleet/pull/12
• Trans transformer block/layer by @FeixLiu in https://github.com/PaddlePaddle/PaddleFleet/pull/11
• [CodeStyle] Ignore PLC0414 in __init__.py files by @SigureMo in https://github.com/PaddlePaddle/PaddleFleet/pull/13
• Complete process_groups_config.py doc and fix typo by @Hz188 in https://github.com/PaddlePaddle/PaddleFleet/pull/15
• [setup] Support source installation of Paddle-Fleet by @risemeup1 in https://github.com/PaddlePaddle/PaddleFleet/pull/14
• improve parallel_state.py for EPHcg & Hcg by @Hz188 in https://github.com/PaddlePaddle/PaddleFleet/pull/16
• [CI] Add approval workflow by @swgu98 in https://github.com/PaddlePaddle/PaddleFleet/pull/10
• support pipeline_parallel schedules by @AlAuAu in https://github.com/PaddlePaddle/PaddleFleet/pull/9
• dev global vars, yaml parse by @Hz188 in https://github.com/PaddlePaddle/PaddleFleet/pull/18
• [CI] fix approval workflow by @swgu98 in https://github.com/PaddlePaddle/PaddleFleet/pull/20
• Add Attention and RoPE classes by @lshpku in https://github.com/PaddlePaddle/PaddleFleet/pull/17
• add mtp by @FeixLiu in https://github.com/PaddlePaddle/PaddleFleet/pull/22
• unittest test_schedules bugfix by @AlAuAu in https://github.com/PaddlePaddle/PaddleFleet/pull/24
• add gpt_model specs by @GuoxiaWang in https://github.com/PaddlePaddle/PaddleFleet/pull/23
• add mlp layer by @blacksheep-Aristotle in https://github.com/PaddlePaddle/PaddleFleet/pull/26
• Trans ln by @FeixLiu in https://github.com/PaddlePaddle/PaddleFleet/pull/25
• Add PackedSeqParams class by @lshpku in https://github.com/PaddlePaddle/PaddleFleet/pull/30
• add psp by @FeixLiu in https://github.com/PaddlePaddle/PaddleFleet/pull/31
• Update LanguageModelEmbedding and add unittest by @lshpku in https://github.com/PaddlePaddle/PaddleFleet/pull/27
• Update RoPE and add unittest by @lshpku in https://github.com/PaddlePaddle/PaddleFleet/pull/28
• [CI] add test workflow by @swgu98 in https://github.com/PaddlePaddle/PaddleFleet/pull/19
• add pipeline/utils.py by @blacksheep-Aristotle in https://github.com/PaddlePaddle/PaddleFleet/pull/33
• [CI] change ut dir by @swgu98 in https://github.com/PaddlePaddle/PaddleFleet/pull/34
• [CI] git name by @swgu98 in https://github.com/PaddlePaddle/PaddleFleet/pull/35
• Placeholder for building transformer configuration parsing by @Hz188 in https://github.com/PaddlePaddle/PaddleFleet/pull/32
• [CI] Add typos white list by @swgu98 in https://github.com/PaddlePaddle/PaddleFleet/pull/36
• change sublayer to sublayer_spec by @FeixLiu in https://github.com/PaddlePaddle/PaddleFleet/pull/39
• fix some test_gpt_model dependencies by @GuoxiaWang in https://github.com/PaddlePaddle/PaddleFleet/pull/40
• Add global Timers for logging by @huangjiyi in https://github.com/PaddlePaddle/PaddleFleet/pull/21
• add check_initialized for dp group by @Hz188 in https://github.com/PaddlePaddle/PaddleFleet/pull/41
• add coverage scripts by @XieYunshen in https://github.com/PaddlePaddle/PaddleFleet/pull/38
• fix coverage scripts by @XieYunshen in https://github.com/PaddlePaddle/PaddleFleet/pull/45
• First successful run of GPTModel model definition by @GuoxiaWang in https://github.com/PaddlePaddle/PaddleFleet/pull/44
• Change fleet core to paddlefleet by @From00 in https://github.com/PaddlePaddle/PaddleFleet/pull/46
• fix coverage bug by @risemeup1 in https://github.com/PaddlePaddle/PaddleFleet/pull/47
• Some fixes to successful run glm4.5 in PaddleFormers by @From00 in https://github.com/PaddlePaddle/PaddleFleet/pull/49
• single card test by @swgu98 in https://github.com/PaddlePaddle/PaddleFleet/pull/50
• mv paddlefleet to src by @risemeup1 in https://github.com/PaddlePaddle/PaddleFleet/pull/52
• use paddle12.6 in single card test by @risemeup1 in https://github.com/PaddlePaddle/PaddleFleet/pull/53
• Add GPTModelEstimator by @huangjiyi in https://github.com/PaddlePaddle/PaddleFleet/pull/59
• Fix attention dim order by @lshpku in https://github.com/PaddlePaddle/PaddleFleet/pull/60
• fix ffn_hidden_size is None when init gpt_mlp by @blacksheep-Aristotle in https://github.com/PaddlePaddle/PaddleFleet/pull/56
• Add estimate_mfu by @huangjiyi in https://github.com/PaddlePaddle/PaddleFleet/pull/62
• refine pyproject.toml by @risemeup1 in https://github.com/PaddlePaddle/PaddleFleet/pull/63
• [CI] add uv pre-commit by @swgu98 in https://github.com/PaddlePaddle/PaddleFleet/pull/65
• [CI] Add nemo megatron approval by @swgu98 in https://github.com/PaddlePaddle/PaddleFleet/pull/51
• Add set_logging and get_logger by @huangjiyi in https://github.com/PaddlePaddle/PaddleFleet/pull/64
• bugfix init GPTModel by @GuoxiaWang in https://github.com/PaddlePaddle/PaddleFleet/pull/54
• [CodeStyle][Ruff] update ruff target-version to py310 by @ooooo-create in https://github.com/PaddlePaddle/PaddleFleet/pull/66
• Supprot custom op by @zhangbo9674 in https://github.com/PaddlePaddle/PaddleFleet/pull/48
• 【lora】fix model for Lora by @xiaoguoguo626807 in https://github.com/PaddlePaddle/PaddleFleet/pull/68
• Ci/multi card config by @swgu98 in https://github.com/PaddlePaddle/PaddleFleet/pull/67
• [CI] add approval for model_parallel_config.py & transformer_config.py by @swgu98 in https://github.com/PaddlePaddle/PaddleFleet/pull/70
• [MoE] Add Base MoE Layer by @hushenwei2000 in https://github.com/PaddlePaddle/PaddleFleet/pull/61
• Delete sharded_state_dict to support FC save/load by @changeyoung98 in https://github.com/PaddlePaddle/PaddleFleet/pull/71
• Align to PaddleFormers by @Waynezee in https://github.com/PaddlePaddle/PaddleFleet/pull/72
• Ignore files generated from uv sync for custom ops by @ooooo-create in...