Skip to content
Navigation Menu
{{ message }}
-
Notifications
You must be signed in to change notification settings - Fork 417
Pull requests: InternLM/xtuner
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Clean up expired objref in replay buffer
#1706
opened Apr 23, 2026 by
YanhuiDua
Collaborator
Loading…
[Refactor] Refactor load spec save planning
#1694
opened Apr 21, 2026 by
HAOCHENYE
Collaborator
Loading…
[WIP] Benchmark rl_design branch logic
#1692
opened Apr 21, 2026 by
hhaAndroid
Collaborator
Loading…
[Refactor] Move _init_load_spec from model __init__ to TrainEngine.build_model
#1676
opened Apr 14, 2026 by
HIT-cwh
Collaborator
Loading…
[Fix] memory leak after the fisrt moe decoder layer
npu
#1630
opened Mar 24, 2026 by
NengXu001
Contributor
Loading…
[Feature] Add SP support for roll_sequence_context in MTP
#1629
opened Mar 24, 2026 by
HAOCHENYE
Collaborator
Loading…
[Refactor] Rename CELossContext to LMHeadLossContext and refactor loss context base class
#1571
opened Mar 12, 2026 by
HAOCHENYE
Collaborator
Loading…
[Feature] Add Multi-Token Prediction (MTP) module implementation
#1570
opened Mar 12, 2026 by
HAOCHENYE
Collaborator
Loading…
[Refactor] Refactor loss context API to support multiple loss types
#1569
opened Mar 12, 2026 by
HAOCHENYE
Collaborator
Loading…
Support customized FSDP AGRS and AGRS Dispatcher
#1554
opened Mar 10, 2026 by
HIT-cwh
Collaborator
Loading…
[Feat] support full graph mode for qwen3.5
#1550
opened Mar 6, 2026 by
CyCle1024
Collaborator
Loading…
[Feature] Layer-wise MoE auxiliary loss (split finalize) and optional async router D2H offload
blocked
#1528
opened Mar 3, 2026 by
tina-wen
Contributor
Loading…
Previous Next
ProTip!
Exclude everything labeled
bug with -label:bug.
You can’t perform that action at this time.
