1. 20 Mar, 2026 1 commit
  2. 19 Mar, 2026 2 commits
  3. 16 Mar, 2026 1 commit
  4. 12 Mar, 2026 2 commits
  5. 11 Mar, 2026 1 commit
  6. 10 Mar, 2026 2 commits
  7. 08 Mar, 2026 1 commit
  8. 07 Mar, 2026 1 commit
  9. 06 Mar, 2026 1 commit
  10. 05 Mar, 2026 2 commits
  11. 03 Mar, 2026 1 commit
  12. 02 Mar, 2026 3 commits
  13. 27 Feb, 2026 1 commit
  14. 26 Feb, 2026 2 commits
  15. 25 Feb, 2026 3 commits
  16. 24 Feb, 2026 3 commits
  17. 20 Feb, 2026 1 commit
  18. 19 Feb, 2026 2 commits
  19. 15 Feb, 2026 1 commit
  20. 13 Feb, 2026 1 commit
  21. 12 Feb, 2026 2 commits
  22. 10 Feb, 2026 2 commits
  23. 09 Feb, 2026 1 commit
    • MatejKosec's avatar
      fix: profiler deployment timeout handling for MoE models (#6086) · 67329d10
      MatejKosec authored
      Wrap wait_for_deployment_ready() in try/except TimeoutError for both prefill and decode profiling sweeps
      On timeout: log error, record via add_profiling_error(), clean up the timed-out deployment, and continue to the next parallelization mapping
      Previously, a single deployment timeout would crash the entire profiler job
      67329d10
  24. 06 Feb, 2026 3 commits