toknX

Distributed inference network for Apple Silicon.

Turn unused Mac compute into tokens you can spend across the network.

Join the network

curl -fsSL https://toknx.co/install-node.sh | bash
View on GitHub
Nodes online 11
Jobs running 6
Tokens generated 101,802,641
Network throughput 184.6 tok/s

Models available now

Live inventory from online contributor nodes.

mlx-community/Llama-3.2-1B-Instruct-4bit 2.4 GB RAM
6 nodes Tier S 1 cr / 1K
mlx-community/Qwen2.5-3B-Instruct-4bit 5.8 GB RAM
4 nodes Tier S 1 cr / 1K
mlx-community/Qwen2.5-Coder-7B-Instruct-4bit 13.6 GB RAM
3 nodes Tier M 2 cr / 1K
mlx-community/Meta-Llama-3.1-8B-Instruct-4bit 15.2 GB RAM
2 nodes Tier M 2 cr / 1K

Live activity

Live

Streaming network events over SSE.

job completed mlx-community/Qwen2.5-Coder-7B-Instruct-4bit • 91 tok
job started mlx-community/Llama-3.2-1B-Instruct-4bit • streaming first tokens
job completed mlx-community/Qwen2.5-3B-Instruct-4bit • 184 tok
job completed mlx-community/Meta-Llama-3.1-8B-Instruct-4bit • 132 tok
job started mlx-community/Qwen2.5-Coder-7B-Instruct-4bit • warming cache

Top contributors 7d

Credits earned from completed jobs.

1st @solinox
Contributor
438.12 credits
2nd @cynotw
Contributor
321.87 credits
3rd @zongrux
Contributor
244.51 credits
4th @stefanz
Contributor
198.34 credits
5th @kariam9k
Contributor
155.09 credits