Skip to content

Pull requests: InternLM/xtuner

New pull request New
85 Open 897 Closed
85 Open 897 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Reduce JsonlDataset memory by using mmap array and use npy to store jsonl meta
#1604 opened Mar 19, 2026 by jayhenry Loading…
16
[Refactor] Reduce memory usage in HardPackDataset via shared memory
#1602 opened Mar 19, 2026 by HAOCHENYE Loading…
7
Support RL VLM of rl_design branch
#1598 opened Mar 19, 2026 by hhaAndroid Loading…
16
Ccy/tf5.2
#1596 opened Mar 18, 2026 by CyCle1024 • Draft
[Improve] Free routed experts ray obj ref to avoid memory leak
#1595 opened Mar 18, 2026 by RangiLyu Loading…
11
【CI】add qwen3.5 cases
#1593 opened Mar 18, 2026 by kkscilife Loading…
support replay buffer save and resume, save_hf in trainer
#1592 opened Mar 18, 2026 by YanhuiDua Loading…
16
[Improve] Optimize ReplayBuffer by using direct method calls
#1591 opened Mar 18, 2026 by RangiLyu Loading…
[Fix] Muon optimizer per-expert orthogonalization for MoE models
#1582 opened Mar 13, 2026 by CyCle1024 Loading…
22
[Feature] Add Multi-Token Prediction (MTP) module implementation
#1572 opened Mar 12, 2026 by HAOCHENYE Loading…
11
[Refactor] Rename CELossContext to LMHeadLossContext and refactor loss context base class
#1571 opened Mar 12, 2026 by HAOCHENYE Loading…
[Feature] Add Multi-Token Prediction (MTP) module implementation
#1570 opened Mar 12, 2026 by HAOCHENYE Loading…
[Refactor] Refactor loss context API to support multiple loss types
#1569 opened Mar 12, 2026 by HAOCHENYE Loading…
Mtp
#1566 opened Mar 12, 2026 by junhaoxjtu Loading…
chore: update CI image to pt29_20260304_35b0bb8
#1564 opened Mar 12, 2026 by CyCle1024 Loading…
[Feature] chunk actor logprob computation for memory saving npu
#1555 opened Mar 10, 2026 by tina-wen Loading…
10
Support customized FSDP AGRS and AGRS Dispatcher
#1554 opened Mar 10, 2026 by HIT-cwh Loading…
10
[Feat] support full graph mode for qwen3.5
#1550 opened Mar 6, 2026 by CyCle1024 Loading…
9
fix claude code workflow
#1540 opened Mar 5, 2026 by HAOCHENYE Loading…
2
[Feature] Domino EP support and training optimizations for InternS1 Pro VL blocked
#1528 opened Mar 3, 2026 by tina-wen Loading…
2
[Optimization] Incremental checkpoint save for dcp on torch 2.7.x (ARM CPU optimization) npu
#1525 opened Mar 3, 2026 by tina-wen Loading…
12
[Feature] Offload optimizer states to CPU to reduce memory
#1524 opened Mar 3, 2026 by tina-wen Loading…
10
Test pr2
#1506 opened Feb 28, 2026 by HAOCHENYE Loading…
10
feat: support fsdp2 muon optimizer
#1486 opened Feb 6, 2026 by RangiLyu Loading…
20
Chunk linear loss
#1473 opened Feb 3, 2026 by jayhenry Loading…
12
ProTip! Follow long discussions with comments:>50.