Observed arrival · 2026-08-31
Ayni wants to turn idle devices into a shared AI utility
A community compute-sharing platform that distributes batch inference across phones, laptops, and GPUs, with quoted prices and device-level payouts.
Field notes
The proposed coordinator turns a batch into parallel device work: it benchmarks hardware, assigns encrypted requests, reassembles results, and meters each contributor’s completed items. The public interface exposes redundancy choices from 1× to 3× and names Qwen2.5 0.5B Instruct as its example model. Its current quote reports no matching devices online, making the page both a technical specification and a visible snapshot of an unproven supply network.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
One card from the complete issue