All pull requests

Pull requests

• Enable multi-rank validation under simulation

  #562·nirvedhmeshram opened Sep 24, 2026Collaborator·

  ·
  8

• chore(deps): bump the python-deps group with 3 updates

  dependenciespython
  #560·dependabot[bot] opened Sep 23, 2026Bot·

  ·

• chore(deps): bump the github-actions group across 1 directory with 8 updates

  dependenciesgithub_actions
  #559·dependabot[bot] opened Sep 23, 2026Bot·

  ·

• Fix stale 08 example path in docs; add pytest coverage for 09 example

  #555·Sophia-Thickums opened Sep 22, 2026·

  ·

• Add an on-demand Iris vs RCCL CCL benchmark job

  #553·nirvedhmeshram opened Sep 22, 2026Collaborator·

  ·
  18

• Test one kernel driving two tensors through two peer tables

  in-progressiris
  #552·mawad-amd opened Sep 22, 2026Collaborator·

  ·
  4

• [experimental] PyTorch symmetric memory as an allocation provider

  #551·nirvedhmeshram opened Sep 17, 2026Collaborator·

  ·
  11

• add multi-GPU Jacobi iteration

  #545·NIne-WIngEd opened Aug 7, 2026·

  ·
  1

  5

• Add iris.concurrent (concurrent, independent, fused comm+gemm)

  #544·ryanswann-amd opened Jul 23, 2026Collaborator·

  ·
  8

• MegaKernels in Iris 🔥

  in-progressiris
  #541·neoblizz opened Jun 24, 2026Member·

  ·
  6

• Support memory sharing in simulation environments

  in-progressiris
  #539·mawad-amd opened Jun 18, 2026Collaborator·

  ·
  4

• Add more tests

  #537·micmelesse opened Jun 17, 2026Contributor·

  ·
  19

• Add vLLM Llama-70B all-reduce reproducer

  #535·micmelesse opened Jun 16, 2026Contributor·

  ·
  4

• Add graph-capturable one-shot all-reduce

  in-progressiris
  #531·mawad-amd opened May 5, 2026Collaborator·

  ·

• Fix matmul_reduce_scatter wrapper passing unsupported bias argument

  in-progressiris
  #530·mawad-amd opened May 2, 2026Collaborator·

  ·
  3

• Fix hardcoded lock value in fused GEMM+CCL operations

  in-progressiris
  #529·mawad-amd opened May 2, 2026Collaborator·

  ·
  1

  7

• Cache hip.get_num_xcc() to eliminate 14ms Config overhead

  in-progressiris
  #528·mawad-amd opened May 2, 2026Collaborator·

  ·
  10

• Add CUDA graph capture probe for iris collectives

  in-progressiris
  #527·mawad-amd opened Apr 30, 2026Collaborator·

  ·
  5

• Validate symmetric heap placement in CCL collectives

  in-progressiris
  #526·mawad-amd opened Apr 30, 2026Collaborator·

  ·
  13

• Add SP benchmarks for GPT-OSS-120B MoE model (GEMM+RS, RS+RMSNorm, E2E)

  in-progressiris
  #513·aamarnat opened Apr 21, 2026Collaborator·

  ·

• Copy Engine AG-GEMM and GEMM-AG benchmarks

  in-progressiris
  #496·dsidler opened Apr 7, 2026Collaborator·

  ·

• Add as_remote(rank) pointer translation API

  in-progressiris
  #495·mawad-amd opened Apr 1, 2026Collaborator·

  ·

• bench: report cross-rank timing stats

  in-progressiris
  #493·mawad-amd opened Apr 1, 2026Collaborator·

  ·
  6

• Add MoEDispatcher for expert-parallel token dispatch/combine

  in-progressiris
  #492·mawad-amd opened Mar 28, 2026Collaborator·

  ·

• Implement ring attention (arXiv:2310.01889)

  #450·mawad-amd with CopilotAI opened Mar 12, 2026·

  ·
  1

  8