SmolCluster: Distributed Training Library for Heterogeneous Hardware Using Raw Sockets
retr0jirachi · x · 2026-09-01
SmolCluster is a distributed deep learning library designed for heterogeneous hardware, built to help users learn distributed training and inference from the ground up. It runs across Mac minis, Raspberry Pis, MacBooks, and Windows machines using only Python's socket library, avoiding dependencies like NCCL or MPI. The library supports various algorithms including FSDP, Classic Data Parallelism, and Model Parallelism. It is explicitly educational, prioritizing code readability over production speed.
More from Infra
- Adani: AI competitive advantage shifts to clean energy and data center integration — Div_pradeep · 2026-09-01
- Naver Proposes Verification-Aware Training to Boost Speculative Decoding Draft Models — naver-ai · 2026-09-01
- ExLlamaV3 update: MoE expert CPU offload, GLM-5.3-Flash, self-calibrated quants — Unstable_Llama · 2026-09-01
- Highlander launches: custom GPU kernels serve realtime video at 50% of competitors' cost — Small-Term672 · 2026-09-01
- Qwen3.8 pMLX Engine: Runs on 12GB RAM at 10 tok/s — EyalToledano · 2026-09-01
- GPU Debt Investors Assume Zero Residual Value and Distrust Spot Pricing — AccBalanced · 2026-09-01