Idle is value — make every device a node of the network
Aggregating idle disk storage, uplink bandwidth and AI compute from around the world, directly connecting storage providers, edge CDNs and compute buyers. No private sale, no public sale — a 1% transaction tax and a 0.5% transfer tax sustain the operations fund, giving the platform real commercial viability.
Five modules form the complete multi-terminal system — click any card to open the corresponding interface
The platform opens up differentiated resource-contribution capabilities according to device form — a core design that sets Guixu apart from ordinary mining-style projects
| Shared resource | PC Client | Mobile App | Notes |
|---|---|---|---|
| Disk storage | Supported · up to 2000 GB | Supported · up to 64 GB | Cold-data archiving and replica redundancy; the larger the capacity, the higher the weight |
| Uplink bandwidth | Supported · up to 480 Mbps | Up to 120 Mbps · Wi-Fi only | Used for edge-cache distribution; unit price rises 15% during night hours |
| AI compute | Supported · up to 14 TFLOPS | Not supported | Mobile devices do not join compute sharing, to protect battery life and thermal control |
| Runtime policies | Idle only / night boost / yield under load | Wi-Fi only / charging only / stop on low battery | Every sharing behaviour can be toggled with one click in the client |
The platform does not rely on token-price narratives — resources connect directly to real demand-side businesses, forming sustainable commercial viability
Fragmented idle space is aggregated into a large-scale cold-data pool, providing replica redundancy and long-term archiving.
Idle uplink spread across regions and carriers is a natural fit for edge-distribution scenarios.
Idle GPU and CPU capacity joins the scheduling pool to serve batch tasks such as model inference and rendering.
The traffic and trading activity accumulated by the platform itself creates a token-economy effect that continuously feeds the operations fund.
Pick the client that suits your device and start contributing idle resources
Framework first, then progressive refinement