-
Notifications
You must be signed in to change notification settings - Fork 276
Pull requests: sgl-project/SpecForge
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
[DataFlow runtime] Phase D — training managers (no_sync, full resume, checkpoint/eval)
#637
opened Jul 1, 2026 by
maocheng23
Collaborator
•
Draft
[DataFlow runtime] Phase C — colocated lightweight control plane
#636
opened Jul 1, 2026 by
maocheng23
Collaborator
•
Draft
[DataFlow runtime] Phase B4 — adopt the de-EAGLE3 surface (cutover + docs + gate)
#635
opened Jul 1, 2026 by
maocheng23
Collaborator
Loading…
Enhance DFlash training details in README
#634
opened Jul 1, 2026 by
catnanami
Contributor
Loading…
6 tasks
[DataFlow runtime] Phase B3 — domain Trainer wrapping the runtime seam
#633
opened Jul 1, 2026 by
maocheng23
Collaborator
Loading…
[DataFlow runtime] Phase B2 — decouple the target engine from the sglang version
#632
opened Jul 1, 2026 by
maocheng23
Collaborator
Loading…
[DataFlow runtime] Phase B1 — TargetEngine ABC + de-EAGLE3 the target boundary
#631
opened Jul 1, 2026 by
maocheng23
Collaborator
Loading…
docs: reconciled SpecForge architecture plan (DataFlow runtime + domain layer)
#630
opened Jun 30, 2026 by
maocheng23
Collaborator
Loading…
[DataFlow runtime] Domino end-to-end + StepContext for schedule-dependent loss
#629
opened Jun 30, 2026 by
maocheng23
Collaborator
Loading…
[DataFlow runtime] DFlash end-to-end on the composable launch (offline + online)
#628
opened Jun 30, 2026 by
maocheng23
Collaborator
Loading…
[DataFlow runtime] Composable launch: StrategySpec registry + parameterized builders
#627
opened Jun 30, 2026 by
maocheng23
Collaborator
Loading…
[DataFlow runtime · online] O1.2 — named builder + interleaved async loop
#625
opened Jun 29, 2026 by
maocheng23
Collaborator
Loading…
[DataFlow runtime · online] O1.1 — shared cross-process control plane
#624
opened Jun 29, 2026 by
maocheng23
Collaborator
Loading…
fix: resolve NPU OOM with default training config
#620
opened Jun 29, 2026 by
curnane-lab
Contributor
Loading…
1 of 6 tasks
feat: add Ascend NPU support for VP-Drafter
#619
opened Jun 29, 2026 by
curnane-lab
Contributor
Loading…
6 tasks
feat(dspark): add Ascend NPU support for Qwen3.5-4B DSpark training
#617
opened Jun 29, 2026 by
curnane-lab
Contributor
Loading…
5 of 6 tasks
feat: DSpark trainer (DFlash + Markov/confidence heads + L1 distillation)
#613
opened Jun 28, 2026 by
maocheng23
Collaborator
Loading…
feat: support training DFlash for Kimi K2.7 Code and Qwen 3.6 27B
#593
opened Jun 24, 2026 by
Boreas618
Loading…
6 tasks
feat: add --save-total-limit to auto-cleanup old checkpoints
#590
opened Jun 23, 2026 by
jxiaof
Loading…
3 of 6 tasks
Expose flex_attention kernel options in DFlash and Domino training
#586
opened Jun 18, 2026 by
heiheiha798
Loading…
[Feature] VLM DFlash Training: Multi-Model Support for Qwen3-VL / Qwen3.5 / Qwen3.6
#585
opened Jun 18, 2026 by
zyk42
Loading…
feat: Train-Inference Disaggregation for Remote Target Model Serving
#573
opened Jun 2, 2026 by
moehanabi
Contributor
Loading…
4 of 6 tasks
Previous Next
ProTip!
Add no:assignee to see everything that’s not assigned.