2 Pull requests merged by 1 person

• Fix diloco test for pt stable

  #204 merged May 30, 2025

• Fix DiLoCo with DTensor

  #197 merged May 27, 2025

2 Pull requests opened by 2 people

• [WIP] Streaming DiLoCo prototype

  #203 opened May 28, 2025

• Generalize DiLoCo to support Streaming

  #205 opened May 30, 2025

4 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• Reuse Valid Accumulated Gradients Upon Failure

  #192 commented on May 26, 2025 • 0 new comments

• Explain quorum

  #163 commented on May 27, 2025 • 0 new comments

• Support multiple quorums on a single LighthouseServer using gRPC metadata-based room assignment

  #189 commented on May 29, 2025 • 0 new comments

• Added example training scripts for localsgd, DiLoCo, Live Checkpoint Recovery, and proactive failure detection with DDP (#198)

  #200 commented on May 25, 2025 • 0 new comments