-
Notifications
You must be signed in to change notification settings - Fork 79
Pull requests: jd-opensource/xllm
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
bugfix: fix memory growth caused by brpc arena configuration.
#409
opened Nov 19, 2025 by
wly-115
Loading…
bugfix: fix memory growth caused by brpc arena configuration.
#408
opened Nov 19, 2025 by
wly-115
Loading…
feat: support Qwen2-VL & GME-Qwen2-VL model on npu device.
#399
opened Nov 18, 2025 by
xanecdotex
Loading…
feat: remove redundant input parameters by add batch forward type.
#396
opened Nov 17, 2025 by
RobbieLeung
Loading…
feat: revert the original code before refactoring the multi-stream.
#395
opened Nov 17, 2025 by
yq33victor
Loading…
refactor: remove embedding allocater of speculative worker.
#394
opened Nov 17, 2025 by
RobbieLeung
Loading…
refactor: optimize vlm_master/mm_handler/message modules.
#392
opened Nov 17, 2025 by
xiao-yu-chen
Loading…
feat: support dp load balance associated with custom mla op implementation.
#389
opened Nov 17, 2025 by
guojinrong-nn
Loading…
feat: support multithread load model weights file.
#388
opened Nov 17, 2025 by
Clement-Wang26
Loading…
feat: enable torch_npu graph mode for Qwen-3 dense with TP support.
#325
opened Nov 6, 2025 by
yingxudeng
Loading…
ProTip!
Filter pull requests by the default branch with base:main.