Skip to content

Pull requests: runwangdl/TrainDeeploy

New pull request New
9 Open 45 Closed
9 Open 45 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

test(gap9): add the LoRA and QLoRA CCT training models to CI
#54 opened Jul 28, 2026 by runwangdl Owner Loading…
feat(PULPOpen): matmul that dequantises an int8 weight in-kernel
#53 opened Jul 28, 2026 by runwangdl Owner • Draft
1
perf(PULPOpen): add a vector bias without materialising the broadcast
#51 opened Jul 28, 2026 by runwangdl Owner Loading…
perf(PULPOpen): let Gemm take a one-dimensional bias instead of widening it
#50 opened Jul 28, 2026 by runwangdl Owner Loading…
Feat/gap9 training power collection
#41 opened Jun 29, 2026 by runwangdl Owner Loading…
5 tasks
Tooling/gap9 cluster hang trace
#35 opened Jun 11, 2026 by runwangdl Owner Loading…
5 tasks
feat: add SpeechNet (SilentWear) training test
#31 opened May 18, 2026 by runwangdl Owner • Draft
1 of 3 tasks
feat(training): untiled L3 baseline — per-step cycle counts for all 4 L3 models
#21 opened May 10, 2026 by runwangdl Owner Loading…
feat(platform): port Siracusa+RedMulE from pulp-platform/Deeploy#67
#20 opened May 10, 2026 by runwangdl Owner Loading…
1 task
ProTip! Add no:assignee to see everything that’s not assigned.