-
Notifications
You must be signed in to change notification settings - Fork 99
Pull requests: ByteDance-Seed/VeOmni
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
[dist] fix: Fall back to legacy gradient divide factor API for EP-FSDP modules due to torch 2.7 compatibility issues.
#205
opened Nov 20, 2025 by
Luosuu
Loading…
[data] feat: add plugin-style custom dataset registry
#203
opened Nov 20, 2025 by
TimYangst
Loading…
6 tasks done
Added related yaml for qwen3 dense model sft
#199
opened Nov 17, 2025 by
A1waysBeenHere
Loading…
4 of 6 tasks
helper: degrade veomni_patch functions to warnings/no-op
#197
opened Nov 14, 2025 by
iqiancheng
Loading…
helper: degrade veomni_patch functions to warnings/no-op
#196
opened Nov 14, 2025 by
iqiancheng
Loading…
Add TensorBoard support for training metrics logging
#195
opened Nov 14, 2025 by
iqiancheng
Loading…
train qwen3-vl-moe on ShareGPT4V-small with quick-start
#194
opened Nov 14, 2025 by
iqiancheng
Loading…
feat: distributed checkpointer support customized backend
#182
opened Nov 8, 2025 by
Ziyi-Wang
Loading…
Optimize Qwen3-Moe Performance on Ascend NPU with Fused Operators & Patches
#167
opened Nov 3, 2025 by
zhihaofang1017
Loading…
[misc] feat: update uv support for aarch platform for Ascend+Kunpeng …
#148
opened Oct 19, 2025 by
pjgao
Loading…
6 tasks done
Bug: Training script with rm-pad-id will only train on subset of dataset.
#17
opened Jun 4, 2025 by
TueVNguyen
Loading…
ProTip!
Add no:assignee to see everything that’s not assigned.