Skip to content

Pull requests: hw-native-sys/pypto-lib

New pull request New
23 Open 656 Closed
23 Open 656 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

Add Qwen3 device top-k candidate selector
#769 opened Jul 13, 2026 by zmnobug Loading…
5
PoC: fold dsv4/moe dispatch/combine wait into syncall barriers enhancement New feature or request
#768 opened Jul 13, 2026 by lwDavid Contributor • Draft
@lwDavid
4
refactor(dsv4): port #734 rope_cs collapse to HCA/SWA decode sparse kernels
#763 opened Jul 13, 2026 by wangqin1723-max Contributor Loading…
4
perf(qwen3-14b decode): remove redundant SPMD task-id dependencies
#762 opened Jul 13, 2026 by Little-oil Loading…
25
Optimize Qwen3 A8W8 decode attention scheduling
#760 opened Jul 13, 2026 by vegetabledoww Contributor Loading…
8
Add direct CANN FAI decode attention for Qwen3
#755 opened Jul 12, 2026 by ChaoWao Contributor Loading…
10
fix(dsv4): zero inactive prefill MoE tails
#744 opened Jul 10, 2026 by high-cloud Contributor Loading…
3
perf(dsv4/moe): launch routed expert stages once per expert
#738 opened Jul 9, 2026 by zhangqi-chen Collaborator • Draft
1
Merge DSV4 compressor ratio modules
#730 opened Jul 8, 2026 by zhaozhaozz Contributor Loading…
12
Fix Qwen3 attention mask NaNs
#714 opened Jul 7, 2026 by superxf Contributor Loading…
1
6
Add device-side LM head to decode fwd
#711 opened Jul 7, 2026 by sjduan Contributor Loading…
2
perf(qwen3-14b decode): early-dispatch infra + critical-path restructuring
#691 opened Jul 6, 2026 by ChaoWao Contributor Loading…
18
refactor(qwen3): drop dead hidden_states arg from decode_fwd
#690 opened Jul 6, 2026 by lyfne123 Contributor Loading…
1
perf(dsv4): qproj wq_b transpose + serving contract; sparse_attn/indexer tuning
#676 opened Jul 2, 2026 by Hzfengsy Member Loading…
6
perf(dsv4 qkv): precompute q-head rope factors
#648 opened Jun 30, 2026 by wangqin1723-max Contributor Loading…
4
perf(dsv4 prefill): avoid indexer pool token scan
#618 opened Jun 25, 2026 by zhaozhaozz Contributor Loading…
2
Add: vector (AIV) row_sum variant for qwen3-14b K/V projection
#586 opened Jun 23, 2026 by Inspiron-st Collaborator Loading…
7
Refactor: scope dsv4 decode+prefill kernels under auto_scope=False
#585 opened Jun 23, 2026 by zhangqi-chen Collaborator Loading…
9
feat(deepseek/v4): gather-free split-half RoPE for the decode path (draft, decode-only) enhancement New feature or request
#564 opened Jun 20, 2026 by Hzfengsy Member • Draft
@lwDavid
4
feat(world_model): Add T5, CLIP, Vae, DiT pypto3.0 kernels
#520 opened Jun 12, 2026 by xzhxzhxzh123 Collaborator Loading…
12
feat(step3p5): initial in-tree drop of step3p5 decoder + Phase A split-scope refactor
#510 opened Jun 11, 2026 by csy0225 • Draft
1 of 4 tasks
15
Add Cayley-Hamilton triangular inverse example
#357 opened May 22, 2026 by ChristosMatzoros Loading…
5
Add: cross-class @pl.jit.inline example
#239 opened May 9, 2026 by bumble0918 Contributor Loading…
5
ProTip! Type g p on any issue or pull request to go back to the pull request listing page.