-
Notifications
You must be signed in to change notification settings - Fork 17
Pull requests: Infini-AI-Lab/astraflow
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
MoE post-training with Rollout Routing Replay (R3), validated on Qwen3-30B-A3B (MATH-500 86.95 -> 92.35)
#24
opened Sep 3, 2026 by
iamziyuzhao
Loading…
feat(lora): fix LoRA weight-sync deadlock and add stable LoRA RL recipe
stale
#20
opened Jun 18, 2026 by
WWWjiahui
Contributor
Loading…
Add Qwen3.5-4B math RL recipes (full + delta) + Qwen3.5 enablement
stale
#11
opened Jun 2, 2026 by
iamziyuzhao
Loading…
ProTip!
Exclude everything labeled
bug with -label:bug.