Edge0-AI

Edge0-AI/Edge0

Run 35B MoE models in ~2.5 GB of RAM. Weights stream from SSD; a trained router predicts loads a step ahead. Mac, iPhone, Android, Windows supported

5.0K stars 888 forks 5.0K watchers Python Apache License 2.0
View on GitHub
3 Open Issues Need Help Last updated: Oct 15, 2026

Open Issues Need Help

View All on GitHub
Benchmark: iOS — on-device benchmark via Edge0PhoneProbe that prints the submission block 13 days ago
enhancement good first issue
Edge0-AI/Edge0
5.0K

Run 35B MoE models in ~2.5 GB of RAM. Weights stream from SSD; a trained router predicts loads a step ahead. Mac, iPhone, Android, Windows supported

Python
Benchmark: Android — on-device benchmark producing the community submission block 21 days ago
enhancement good first issue
Edge0-AI/Edge0
5.0K

Run 35B MoE models in ~2.5 GB of RAM. Weights stream from SSD; a trained router predicts loads a step ahead. Mac, iPhone, Android, Windows supported

Python
Benchmark: Windows — one-command script that prints the community submission block 21 days ago
enhancement good first issue
Edge0-AI/Edge0
5.0K

Run 35B MoE models in ~2.5 GB of RAM. Weights stream from SSD; a trained router predicts loads a step ahead. Mac, iPhone, Android, Windows supported

Python