Skip to content

Pull requests: InfiniTensor/InfiniOps

New pull request New
39 Open 600 Closed
39 Open 600 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

feat(operator):support nvidia-silu-operator.
#647 opened Jun 5, 2026 by pengcheng888 Loading…
78 tasks
feat(operator):support nvidia-embedding-operator.
#646 opened Jun 5, 2026 by pengcheng888 Loading…
2 of 78 tasks
Feat/metax cuda add rms norm operator
#645 opened Jun 5, 2026 by pengcheng888 • Draft
2 of 78 tasks
1
refactor InfiniOps cpu runtime through InfiniRT
#644 opened Jun 4, 2026 by spike-zhu Loading…
78 tasks
build: record physical plugin split experiment
#643 opened Jun 4, 2026 by voltjia Collaborator • Draft
1
build: read plugin CMake entry from manifest
#642 opened Jun 4, 2026 by voltjia Collaborator • Draft
ci: make shadow workflow plugin-aware
#641 opened Jun 4, 2026 by voltjia Collaborator • Draft
build: support external plugin roots
#640 opened Jun 4, 2026 by voltjia Collaborator • Draft
ci: add plugin test matrix helper
#639 opened Jun 4, 2026 by voltjia Collaborator • Draft
docs: document plugin contract
#638 opened Jun 4, 2026 by voltjia Collaborator • Draft
feat: add local build/test workflows and operator report tooling
#637 opened Jun 4, 2026 by gongchensu Contributor Loading…
78 tasks
feat(operator):support Cuda-AddRmsNorm-operator.
#635 opened Jun 3, 2026 by pengcheng888 Loading…
2 of 78 tasks
2
Add build-time plugin contract
#633 opened Jun 3, 2026 by voltjia Collaborator • Draft
fix(cuda/gemm): make caches safe for threaded callers
#632 opened Jun 3, 2026 by voltjia Collaborator Loading…
46 of 64 tasks
1
1
ci: fix platform-specific builds
#630 opened Jun 2, 2026 by zhangyue207 Collaborator • Draft
32 of 44 tasks
test(torch): skip high-memory cuda svd reference
#628 opened May 29, 2026 by zhangyue207 Collaborator • Draft
26 of 44 tasks
fix(ninetoothed): bound RmsNorm autotune shapes
#624 opened May 28, 2026 by voltjia Collaborator • Draft
29 of 44 tasks
feat: build C API on functional operator API
#617 opened May 20, 2026 by voltjia Collaborator • Draft
53 tasks done
6
feat(ascend): add top-k top-p sampler
#603 opened May 12, 2026 by zhangyue207 Collaborator • Draft
52 of 69 tasks
build(ascend): support custom kernel builds
#602 opened May 12, 2026 by zhangyue207 Collaborator • Draft
34 of 49 tasks
feat(bindings): support scalar default wrapper generation
#601 opened May 12, 2026 by zhangyue207 Collaborator • Draft
36 of 51 tasks
feat(ascend): add scaled softmax operator
#599 opened May 12, 2026 by zhangyue207 Collaborator • Draft
45 of 61 tasks
feat(ascend): add embedding operator
#598 opened May 12, 2026 by zhangyue207 Collaborator • Draft
45 of 61 tasks
feat(ascend): op-norm-rope group (re-PR after #72 revert) — Swiglu, SiluAndMul, CausalSoftmax, RmsNorm, AddRmsNorm, RotaryEmbedding
#73 opened Apr 25, 2026 by zhangyue207 Collaborator Loading…
58 of 73 tasks
1
feat: Add cpu random sample op
#68 opened Apr 21, 2026 by baominghelly • Draft
@baominghelly
ProTip! Exclude everything labeled bug with -label:bug.