Kernels
optimizer / test

Commit History

Apply pre-commit formatting (isort) [skip-build]
96b287c

wyldecat Claude Opus 4.6 commited on

Add MoE uneven shard test with mixed expert and non-expert params [skip-build]
bdada12

wyldecat Claude Opus 4.6 commited on

Add uneven shard correctness test [skip-build]
1a97671

wyldecat Claude Opus 4.6 commited on

Update tests for MoE and parallel optimizations [skip-build]
81f49fe

wyldecat Claude Opus 4.6 commited on

Add torch.compile, CUDA graph, and compiled momentum [skip-build]
e74d98f

wyldecat Claude Opus 4.6 commited on

Apply suggestions from code review
cdaaf4f

TaehyunKim Copilot commited on

Support multi-component expert_keys (e.g. "experts.w1")
5a99e12

wyldecat Claude Opus 4.6 commited on

Use component-level matching for expert_keys to avoid shared_experts collision
f008017

wyldecat Claude Opus 4.6 commited on

Normalize parameter FQNs to handle torch.compile / checkpoint wrappers
95a620f

wyldecat Claude Opus 4.6 commited on

Refactor pipeline to async generator pattern (#16)
33929c0
unverified

wyldecat github-actions[bot] commited on

Support mHC (#15)
ae32572
unverified

wyldecat github-actions[bot] commited on

Support param group with various placements (#13)
e2b41e5
unverified

wyldecat github-actions[bot] commited on

misc
35894d1

ca1207 commited on

apply all2all scatter gather
ff6d675

ca1207 commited on

feat(muon_clip) : add muon clip (#6)
d65066c
unverified

dongseokmotif dongseokmotif github-actions[bot] commited on

feat: update muon to receive paramgroups, not model (#4)
b0f46c7
unverified

leejunhyeok junhyeok.lee wyldecat commited on

feat(muon): add test for muon
3261444

wyldecat commited on