All pull requests

Pull requests

• docs: fix 22 typos in documentation

  #519·Avicennasis opened Jun 13, 2026·

  ·
  1

• [compiler]Implement Triton backend features and fix related bugs

  #513·6somehow opened Jul 25, 2025Contributor·

  ·

• [torch-frontend] use new register method to register byteir.flash_att…

  #473·qingyunqu opened Oct 25, 2024Collaborator·

  ·

• fix slice opt returns copy

  #418·YellowHCH opened Aug 7, 2024Collaborator·

  ·

• [compiler] Horizontal fusion

  #395·YellowHCH opened Jul 8, 2024Collaborator·

  ·
  3

• [compiler] add forall-tiling pass

  enhancement
  #370·XG-zheng opened Jun 24, 2024Collaborator·

  ·

• [compiler] fold mhlo.reduce with splat input

  #367·jianwenyyy opened Jun 24, 2024Collaborator·

  ·

• [WIP] Add multi stream execute plan

  #340·Xinyu302 opened Jun 16, 2024·

  ·
  1

• [compiler] try to reorder byre.copy to overlap IO and compute

  #332·YellowHCH opened Jun 13, 2024Collaborator·

  ·

• [dynamo] reduce host side overhead

  #328·YellowHCH opened Jun 12, 2024Collaborator·

  ·
  1

• [GEMM codegen] Add e2e pipeline

  #316·Xinyu302 opened Jun 7, 2024·

  ·
  1

• [runtime] add BatchTranspose cuda kernel

  enhancement
  #309·XG-zheng opened Jun 5, 2024Collaborator·

  ·

• [torch-frontend] fx export FakeTensor weight as SplatElementsAttr.

  #299·Vremold opened May 30, 2024Collaborator·

  ·

• Training

  #296·YellowHCH opened May 30, 2024Collaborator·

  ·

• [fix] Update prepare.sh to prevent deletion of 'external_libs/libs' before rebuilding runtime

  #270·Xinyu302 opened May 22, 2024·

  ·
  1

• [WIP][E2E]Fix e2e errors

  #258·Xinyu302 opened May 18, 2024·

  ·
  1

• [torch-frontend] add stablehlo IRs for Mixtral model.

  enhancement
  #254·Vremold opened May 16, 2024Collaborator·

  ·
  1

• [compiler] support shape reification for callOp

  enhancement
  #182·XG-zheng opened Apr 7, 2024Collaborator·

  ·

• [Runtime]Add runtime dynamic shared mem support

  enhancement
  #179·Xinyu302 opened Apr 2, 2024·

  ·
  1

• WIP: [compiler] implement reduction with gpu shuffle

  enhancement
  #171·Mi-Jiazhi opened Mar 28, 2024Contributor·

  ·

• [draft][runtime] modified OpKernel to support dynamic shape

  enhancement
  #162·XG-zheng opened Mar 25, 2024Collaborator·

  ·

• [Doc] Update gpu.md

  documentation
  #8·lchang20 opened Feb 20, 2023·

  ·
  3