Nodexo Crosses 50 Verified GPUs on Bittensor
Nodexo, Bittensor subnet 106, passes 50 verified GPUs across 7 classes and 6 countries with 99.9% 24-hour reliability and hourly USDC rentals.
Nodexo, operating as Bittensor subnet 106, has recently achieved a significant milestone by surpassing 50 verified GPUs online. The network features a diverse range of GPU classes from six countries, encompassing both consumer-grade RTX 4090s and high-end NVIDIA H200s. With an impressive reliability rate of 99.9 percent over a 24-hour period, Nodexo ensures continuous GPU monitoring and verification.
Each GPU is readily available for rental using a single command, offering transparency and accessibility within the Bittensor ecosystem. The network employs proof-of-compute techniques and hardware attestation to validate hardware and establish reliability scores. Rental demand takes precedence, while idle machines can be utilized for managed training workloads to maintain optimal utilization.
Payments are processed hourly in USDC, and accountless rentals are possible through agent-compatible interfaces. The current NVIDIA hardware mix supports both inference and training workloads. Nodexo's operators oversee inventory, monitor reliability metrics, provision access via command-line tools, and facilitate connections through standard SSH once a rental commences.
This achievement comes at a time when there is increasing interest in decentralized alternatives to centralized cloud providers, and Nodexo aims to enhance high-performance compute accessibility and transparency. The live capacity of the network continuously evolves as machines join, move between free, rented, and audit states, and are redeployed.
Nodexo positions itself as a trust layer for decentralized GPU resources as the network continues to expand. Developers and operators can track inventory and reliability figures on Nodexo's dedicated statistics page.
Written by urgent.news from HackerNoon's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
