Grove: Distributed ML Training over AirDrop
swar_ja
32 points
1 comment
March 25, 2026
Related Discussions
Found 5 related stories in 82.3ms across 8,303 title embeddings via pgvector HNSW
- Decoupled DiLoCo: Resilient, Distributed AI Training at Scale metadat · 45 pts · April 27, 2026 · 52% similar
- Grok 4.3 simianwords · 386 pts · May 01, 2026 · 49% similar
- Elon Musk confirms xAI used OpenAI's models to train Grok fraXis · 12 pts · April 30, 2026 · 47% similar
- Grok Build meetpateltech · 88 pts · May 14, 2026 · 47% similar
- Language model teams as distributed systems jryio · 87 pts · March 16, 2026 · 46% similar
Discussion Highlights (1 comments)
swar_ja
Python library for training models across MacBooks over wireless. Uses AWDL for discovery and fallback transport, upgrades to WiFi when available. Works on networks with client isolation (eduroam etc.). grove start train.py -n 4 on one Mac, grove join on the others. DiLoCo for communication-efficient sync, optional SparseLoCo compression (~30x) for slower links. Swift sidecar for the AWDL layer since Network.framework isn't accessible from Python. macOS/Apple Silicon only, built on MLX.