Skip to content

Pull requests: jd-opensource/xllm

New pull request New
20 Open 439 Closed
20 Open 439 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

feat: support multiple separate atb workspaces
#552 opened Dec 16, 2025 by zhang-minchao Loading…
bugfix: support PD for MTP.
#551 opened Dec 16, 2025 by RobbieLeung Loading…
bugfix: use double buffer to ensure there is no conflict between step and output handling.
#549 opened Dec 16, 2025 by yq33victor Loading…
feat: concurrent multi stream executor for rec_model
#548 opened Dec 15, 2025 by zhang-minchao Loading…
2
feat: adapt host block manager for disagg PD.
#546 opened Dec 15, 2025 by Kang-Meng Loading…
feat: add service guard to reject invalued requests.
#545 opened Dec 15, 2025 by RobbieLeung Loading…
bugfix: fix round fail of LatencySchedule UT in a3.
#544 opened Dec 15, 2025 by weizhehuang0827 Loading…
1
refactor: refactor mm_data module.
#539 opened Dec 13, 2025 by xiao-yu-chen Loading…
5
feat: initialize flashinfer planinfo at layer-0 forward stage.
#534 opened Dec 12, 2025 by yq33victor Loading…
feat: add a new manual management weights loader style for deepseek.
#531 opened Dec 12, 2025 by Clement-Wang26 Loading…
feat: support custom shared memory size.
#528 opened Dec 12, 2025 by Clement-Wang26 Loading…
2
feat: support graph executor on mlu device.
#526 opened Dec 12, 2025 by a120092009 Loading…
11
feat: add mm embed worker and service.
#518 opened Dec 11, 2025 by dongxianzhe Loading…
1
refactor: remove empty_kv_cache and global_empty_kv_cache.
#514 opened Dec 10, 2025 by RobbieLeung Loading…
1
feat: vlm support binary mm input.
#507 opened Dec 9, 2025 by xiao-yu-chen Loading…
3
bugfix: fix the issue of ineffective input embedding transmission.
#490 opened Dec 5, 2025 by magicheng0816 Loading…
7
refactor: optimize unique token count preparation of batch input builder.
#449 opened Nov 27, 2025 by RobbieLeung Loading…
feat: support Qwen2-VL & GME-Qwen2-VL model on npu device.
#399 opened Nov 18, 2025 by xanecdotex Loading…
11
feat: enable torch_npu graph mode for Qwen-3 dense with TP support.
#325 opened Nov 6, 2025 by yingxudeng Loading…
21
【WIP】feat: add rec framwork.
#305 opened Oct 31, 2025 by DragonFive Loading…
3 of 5 tasks
2
ProTip! Adding no:label will show everything without a label.