Skip to content

Pull requests: Tencent/hpc-ops

New pull request New
10 Open 37 Closed
10 Open 37 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

Improve source build robustness
#78 opened Jul 31, 2026 by liuyao0322 Loading…
Fix AllReduce benchmark README script name
#77 opened Jul 31, 2026 by liuyao0322 Loading…
[ROCm] Add ROCm support and norm op.
#69 opened Jul 20, 2026 by WhatGhost Loading…
1
perf: add small-batch route MMA for MoE
#68 opened Jul 20, 2026 by Religious-J Contributor Loading…
feat: add qwen3 tts rope operator
#67 opened Jul 17, 2026 by l-wave Loading…
fix: change kHalfFltEps from 1e-20 to 5e-8 in sampler rng, aligned to torch
#65 opened Jul 13, 2026 by LINZHI97 Loading…
feat: add qwen3 tts topk gumbel sampler
#64 opened Jul 10, 2026 by l-wave Loading…
feat: relax heads_per_group constraint to support 1~8 for BF16 and FP8 decode attention
#58 opened Jun 25, 2026 by Religious-J Contributor Loading…
feat: bf16 fused moe kernel
#41 opened Apr 9, 2026 by ZelinMa557 Loading…
5
feat: add bf16 group gemm kernel
#34 opened Mar 23, 2026 by ZelinMa557 Loading…
ProTip! Add no:assignee to see everything that’s not assigned.