All pull requests

Pull requests

• feat(audio): add TTS Granary annotation pipeline from tn_raw

  #2442·karpnv opened Sep 22, 2026·

  ·
  1

• [benchmarking] Use source-tree benchmark setup

  #2441·rlratzel opened Sep 22, 2026·

  ·
  2

• fix(interleaved): handle null WDS content types

  r1.4.0
  #2438·VibhuJawa opened Sep 22, 2026·

  ·
  3

• docs: draft 26.09 release and migration notes

  #2433·lbliii opened Sep 21, 2026·

  ·
  2

  

• docs: document audio transcript processing

  #2432·lbliii opened Sep 21, 2026·

  ·
  2

  

• docs: document ASR adapter selection

  #2431·lbliii opened Sep 21, 2026·

  ·
  2

  

• feat: support create mode in DataDesignerStage

  community-request
  #2430·Abdullah49645 opened Sep 19, 2026·

  ·
  1

  8

• Fuzzy deduplication evaluation tutorial

  #2426·sarahyurick opened Sep 18, 2026·

  ·
  1

• feat(dedup): support batch_size/process_batch in MinHashStage

  community-request
  #2423·Lingikaushikreddy opened Sep 17, 2026·

  ·
  1

  9

• fix(serve): support preinstalled Dynamo environments

  #2422·praateekmahajan opened Sep 17, 2026·

  ·
  1

• fix(stages): floor gpu_memory_gb-derived GPU fraction at 0.1 so it ne…

  #2421·satishra-ai opened Sep 17, 2026·

  ·
  3

• docs(tutorials): add exact deduplication end-to-end notebook

  community-requestwaiting-on-maintainers
  #2420·chethanuk opened Sep 16, 2026·

  ·
  1

  3

• fix(core): add opt-in cleanup of RayClient's Ray session dir

  community-requestwaiting-on-maintainers
  #2419·chethanuk opened Sep 16, 2026·

  ·
  1

  7

• feat: allow outputs() to declare a spec per input task type

  community-requestwaiting-on-maintainers
  #2418·chethanuk opened Sep 16, 2026·

  ·
  1

  4

• feat(download): write Common Crawl downloads to an fsspec download_dir

  community-requestwaiting-on-maintainers
  #2417·chethanuk opened Sep 16, 2026·

  ·
  1

  4

• docs: document resumability preconditions and guard unsupported checkpoint filesystems

  #2416·abhinavg4 opened Sep 16, 2026·

  ·
  6

• docs: document Lance readers and writers

  docs-only
  #2415·lbliii opened Sep 16, 2026·

  ·
  15

  

• docs: document sound event detection

  docs-only
  #2414·lbliii opened Sep 16, 2026·

  ·
  3

  

• docs: document Ray Data scheduler diagnostics

  docs-only
  #2413·lbliii opened Sep 16, 2026·

  ·
  5

  

• docs: document deduplication controls

  docs-only
  #2412·lbliii opened Sep 16, 2026·

  ·
  9

  

• docs: update ProcessingStage input contract

  docs-only
  #2411·lbliii opened Sep 16, 2026·

  ·
  5

  

• [Do not Merge] build: add Nemotron OCR extra

  #2407·VibhuJawa opened Sep 16, 2026·

  ·
  6

  

• fix: bound Ray Serve startup

  #2406·VibhuJawa opened Sep 16, 2026·

  ·
  6

  

• fix: make file discovery failures actionable

  #2404·VibhuJawa opened Sep 16, 2026·

  ·
  4

  

• Add benchmark script for LLMJudgeWorkflow

  #2401·sarahyurick opened Sep 15, 2026·

  ·
  1