OpenCompute
A decentralized compute layer that aggregates idle GPUs/CPUs into a global network.
Zero-Setup, Scalable, and Developer-Friendly
OpenCompute delivers a seamless and radically simplified compute onboarding experience. Unlike traditional decentralized compute networks that require complex hardware configurations and software setups, OpenCompute introduces a breakthrough: browser-native compute activation. No installations. No environment tuning. Just sign in, and your device becomes part of a global distributed compute network.
This zero-barrier access transforms everyday devices into usable compute nodes—instantly and elastically—becoming a foundational pillar for OpenCompute’s scalable compute infrastructure.
From the developer’s perspective, the OpenCompute SDK offers true plug-and-play usability. With just a few lines of Python—no complex dependencies or migrations—users can seamlessly trigger powerful remote compute tasks.
We’re also launching remote GPU scheduling, designed with a minimalist interface. For example:
RemoteGPUAdd(1, 2)
This is high-performance computing distilled into a single line of code—delivering real-time responsiveness, on-demand scalability, and massive productivity gains for AI, science, and engineering workloads.
OpenCompute redefines compute usability: from “setup-intensive” to “instant-ready.”
Roles in OpenCompute
Users Uses the Python SDK to define and trigger remote functions ("runes") as if calling a local async method.
Worker Nodes Provide compute power (CPU/GPU) to execute assigned tasks. Rewards are distributed in esNEB based on hardware capacity, uptime, task completion, and reputation.
Verifier Nodes Validate Worker outputs to ensure accuracy and network security. They also handle scheduling, synchronization, and support Delegation, where staked esNEB boosts their rewards and ranking.
Delegators Support Verifier Nodes by staking esNEB, participating in governance, and sharing in validation rewards — without needing to run hardware.
Last updated