Stardock adds Hivemind to Clairvoyance to share idle GPUs

Nov 6

1. 02:00

   Unite.AI

   Stardock Clairvoyance 0.90 adds Hivemind, enabling shared local inference on companies' idle GPUs

   Stardock has added Hivemind to the Locama local model manager in Clairvoyance 0.90, allowing team members to share local inference tasks using idle GPUs within the company. The version was released on November 5 and supports Windows, macOS, and Linux. The original article states that Hivemind is peer-to-peer shared local inference, with the Concierge component selecting the model and machine for each request round, and does not describe tensor parallelism or pooled VRAM approaches. The Professional edition costs $20 per month and includes commercial licensing, Hivemind, and team features, while the Enterprise edition costs $100 per seat per month.

   Expand summary