main
[misc] refactor launch API and tensor constructor (#5666)
[Feature] Distributed optimizers: Lamb, Galore, CAME and Adafactor (#5694)
[hotfix] fixed memory usage of shardformer module replacement (#5122)
[misc] update pre-commit and run all files (#4752)
[compatibility] support torch 2.2 (#5875)
[DTensor] refactor sharding spec (#2987)
[Auto Parallel]: Speed up intra-op plan generation by 44% (#5446)