All pull requests

Pull requests

• bugfix: preflight MiniMax FP8 checkpoint indexes.

  #2667·hswsplendid opened Nov 4, 2026Contributor·8 of 17 tasks·

  ·
  4

• bugfix: validate MiniMax FP8 dequantization inputs.

  #2665·hswsplendid opened Nov 4, 2026Contributor·8 of 17 tasks·

  ·
  3

• feat: add audio transcription, DCP execution, and linear cache prefetch.

  #2662·xllm-org opened Nov 3, 2026Collaborator·

  ·
  10

• feat: Add standalone T5 QKV append operator.

  #2656·wangzhao-11a opened Nov 3, 2026·10 of 14 tasks·

  ·

• feat: add DualAffineLayerNorm operator.

  #2655·fbossing-123 opened Nov 2, 2026·

  ·
  9

• docs: add DualAffineLayerNorm operator design review

  #2654·fbossing-123 opened Nov 2, 2026·

  ·
  24

• bugfix: reconcile vmm cache usage after batch retirement.

  #2652·AII12754 opened Nov 2, 2026Contributor·2 tasks done·

  ·
  3

• feat: share static physical pages across hybrid caches.

  #2651·liutongxuan opened Nov 2, 2026Collaborator·17 tasks·

  ·
  2

• perf: defer task pipeline input reclamation after forward.

  #2650·liutongxuan opened Nov 2, 2026Collaborator·17 tasks·

  ·
  7

• refactor: remove unsupported task pipeline implementations.

  #2646·liutongxuan opened Nov 2, 2026Collaborator·17 tasks·

  ·
  4

• feat: Add T5 QKV cache append wrapper.

  #2641·brandfucker opened Nov 1, 2026·6 of 8 tasks·

  ·

• refactor: unify decode row construction into decode_input_builder.

  #2640·wai-bao opened Nov 1, 2026Collaborator·

  ·
  1

• feat: enable adaptive qwen mtp verification on npu.

  #2632·hswsplendid opened Oct 31, 2026Contributor·8 of 17 tasks·

  ·
  2

• feat: manage sliding window caches for block drafts.

  #2627·PixelFlat opened Oct 31, 2026Contributor·9 of 10 tasks·

  ·
  9

• bugfix: support official torch npu wheels.

  #2624·XuZhang99 opened Oct 30, 2026Collaborator·6 of 9 tasks·

  ·

• feat: support measured graph costs for adaptive verification.

  #2623·hswsplendid opened Oct 30, 2026Contributor·11 of 17 tasks·

  ·
  1

• refactor: centralize dp rank placement in scheduler.

  #2621·PixelFlat opened Oct 30, 2026Contributor·7 of 12 tasks·

  ·
  8

• refactor: hoist draft weight sharing to the draft-model spec base.

  #2613·wai-bao opened Oct 29, 2026Collaborator·

  ·
  1

• feat: add t5_split_qkv_append_cache npu op

  #2610·wisdom1710 opened Oct 29, 2026·

  ·
  4

• feat(npu): add DualAffineLayerNorm binding

  #2609·wisdom1710 opened Oct 29, 2026·

  ·
  5

• docs: add T5 QKV cache operator design.

  #2604·wangzhao-11a opened Oct 28, 2026·8 of 17 tasks·

  ·
  3

• perf: optimize npu sampling and sparse dflash2 verification.

  #2603·AII12754 opened Oct 28, 2026Contributor·13 of 16 tasks·

  ·
  30

• docs: add Huawei branding to Ascend NPU entry in README and README_zh

  #2585·yzeyu71 opened Oct 26, 2026·

  ·
  3

• perf: optimize NPU Qwen3.5 GDN prefill projections and packing.

  #2577·AII12754 opened Oct 25, 2026Contributor·8 of 13 tasks·

  ·
  4

• perf: fuse qwen python attention output gating.

  #2576·hswsplendid opened Oct 25, 2026Contributor·9 of 17 tasks·

  ·
  12