main
[CI] disable timm_regnetv_040 as aten::_unique2 is not supproted
[misc] update pre-commit and run all files (#4752)
[test] Fix/fix testcase (#5770)
[plugin] support get_grad_norm (#6115)
[shardformer] support pipeline for deepseek v3 and optimize lora save (#6188)
[misc] update torch version (#6206)
[misc] refactor launch API and tensor constructor (#5666)
[ci] update ci (#6254)
[hotfix] fix testcase in test_fx/test_tracer (#5779)
[upgrade]Upgrade transformers (#6320)
[fix] fix_lazy_init for deepseek model in transformers
[FP8] rebase main (#5963)
[fp8] Merge feature/fp8_comm to main branch of Colossalai (#6016)
[hotfix] moe hybrid parallelism benchmark & follow-up fix (#6048)
[CI] Cleanup Dist Optim tests with shared helper funcs (#6125)
[hotfix] fix hybrid checkpointio for sp+dp (#6184)
fix dist log prob test
[inference] Add smmoothquant for llama (#4904)
release
[zero] Update sharded model v2 using sharded param v2 (#323)
[checkpointio] support load-pin overlap (#6177)