# @gajesh: Darkbloom turns idle Macs into a private inference network

Source: https://x.com/gajesh/status/2044067335506669782?s=46
Kiwi News: https://news.kiwistand.com/stories/gajesh-Darkbloom-turns-idle-Macs-into-a-private-inference-network?index=0x69f1d78426829e0df1c29c09e1095a33ebf77c7b625daa52a7cda5f3d6dc5d2de89e1039
Submitted by thatalexpalmer.eth on 2026-04-29 UTC, 2 upvotes, 0 comments

## Summary

Darkbloom is a project that aims to connect idle Macs directly to inference demand, bypassing what its authors call the Inference Tax, the margin layers that they say sit between AI API calls and the hardware. The post says there are over 100 million Macs with Apple Silicon, that a $3k Mac can run a 60B model at 30 watts, and that most sit idle most of the day. The project describes itself as a private inference network for idle Macs, with its paper and code open at darkbloom.dev and invites for free credits offered in replies.

## Article

Wake the world's sleeping compute.

Look at the Mac nearest to you. What's it doing?

Probably nothing.

There are 100M+ Macs with Apple Silicon out there. Apple quietly made them *really* good at inference. A $3k Mac runs a 60B model at 30 watts.

Most sit idle most of the day.

Meanwhile every AI API call passes through three layers of margin before reaching the hardware. We call this the Inference Tax.

We got curious: what happens if you connect idle Macs directly to inference demand?

This is Darkbloom. Private inference network for idle Macs.

darkbloom [dot] dev -- paper + code open.

Reply for invite + free credits ↓