The Worldwide Supercomputer
What if every idle computer in the world could work together?
Right now, billions of devices sit unused — gaming PCs after the session ends, workstations overnight, smartphones charging, smart TVs in standby. Each one has processing power going to waste. OmnibusCloud connects these devices into a global compute mesh — a supercomputer that spans continents.
Vision: A worldwide supercomputer where anyone can contribute resources and anyone can run distributed workloads.
Two Ways to Participate
Contribute & Earn
Install the OmnibusCloud agent on your devices. When they're idle, they join the global network and process tasks. You earn based on:
- Benchmark score — More powerful devices earn more
- Uptime — Consistent availability is rewarded
- Tasks completed — Direct compensation for work done
Your gaming PC could earn while you sleep. Your office computers could generate revenue overnight. Even a smartphone charging on your nightstand contributes — and earns.
Consume & Compute
Need massive computational power? Submit jobs to OmnibusCloud and tap into the global mesh:
- Pay-per-use — Only pay for compute you actually use
- Instant scale — Access thousands of devices immediately
- No infrastructure — No servers to buy, configure, or maintain
A indie game studio can render their trailer on the same scale as a major studio — without the render farm.
Extreme Heterogeneity
Unlike traditional cloud providers with uniform hardware, OmnibusCloud embraces diversity:
| Device Type | Benchmark Score | Role |
|---|---|---|
| Dedicated render servers | ~2000 | Heavy lifting |
| Gaming PCs | 500-1500 | GPU-intensive tasks |
| Workstations | 300-800 | General compute |
| Laptops | 100-400 | Batch processing |
| Smart TVs, set-top boxes | 20-50 | Lightweight tasks |
| Mobile devices | 10-30 | Micro-tasks |
Performance spread: 100×+ — and the system handles it automatically.
A smart TV with a benchmark score of 30 works alongside a dedicated server scoring 3000. The server processes 100× more tasks — but the TV still contributes meaningfully. No manual configuration. No capability tiers. Just measured performance and proportional allocation.
Why Decentralized?
OmnibusCloud's decentralized model offers advantages that traditional data centers cannot match:
Energy Efficiency
Traditional data centers run 24/7, consuming massive amounts of electricity even when idle. OmnibusCloud devices contribute only during their natural idle periods — no additional energy consumption, no new hardware manufacturing. Computing that's sustainable by design.
Unlimited Scalability
With each new device, OmnibusCloud grows. There's no capacity ceiling, no infrastructure bottleneck. The network scales organically as more people contribute.
Global Reach
OmnibusCloud democratizes access to high-performance computing. A researcher in a region without data center infrastructure can tap into the same global mesh as a Silicon Valley startup. Compute becomes location-independent.
Resilience
Nodes join and leave continuously. A gaming PC goes offline when its owner starts playing — the system adapts instantly. No single point of failure. No maintenance windows. The mesh is always available.
The Scale Difference
What does global scale actually mean? Here's a comparison:
| Project | Local (20 nodes) | OmnibusCloud (10,000+ nodes) |
|---|---|---|
| Feature film (130K frames) | 22.5 days | ~1 hour |
| ML hyperparameter search (10K configs) | 5 days | 6 minutes |
| Video transcoding (1000 files) | 50 hours | 6 minutes |
The script doesn't change. Only the scale does.
The Roadmap
OmnibusCloud launches in phases, building the ecosystem step by step:
| Phase | Status | Description |
|---|---|---|
| Phase 1 | 🚀 Coming Soon | Closed beta with Blender rendering |
| Phase 2 | Planned | Open platform for custom plugins |
| Phase 3 | Future | Full ecosystem with marketplace |
Phase 1: Blender Rendering
The initial beta focuses on a single, well-defined use case: distributed Blender rendering.
Why Blender?
- Clear, measurable workload (frames)
- Active community of artists and studios
- Proven demand for render farm alternatives
- Perfect fit for distributed computing
This allows us to validate the platform, refine the user experience, and build a community before opening to broader applications.
Phase 2: Open Platform
Custom plugins enable any distributed workload:
- Scientific simulations
- Machine learning training
- Data processing pipelines
- Video transcoding
- Your custom application
Phase 3: Marketplace
A full ecosystem where:
- Plugin developers sell their extensions
- Compute providers compete on price and reliability
- Job submitters find the best resources for their workloads
How It Works
OmnibusCloud is powered by WitEngine — the same distributed computing framework used in enterprise WitCloud deployments.
When you submit a job:
- Task Segmentation — Your job is split into parallelizable chunks
- Node Matching — Tasks are matched to capable devices worldwide
- Intelligent Distribution — Work is allocated proportionally to benchmark scores
- Fault Tolerance — If a device disconnects, its tasks are automatically reassigned
- Result Aggregation — Completed results are collected and delivered
Intelligent Batching
The key innovation: all nodes finish simultaneously, regardless of their speed.
Each node receives work proportional to its benchmark score. The server processes 100× more frames than the phone — but they both complete at the same time. Maximum resource utilization, minimal idle waiting.
The same script that runs on 10 office machines can run on 10,000 devices worldwide — no changes needed.
Benefits
For Contributors
| Benefit | Description |
|---|---|
| Passive income | Earn while your devices sit idle |
| Zero effort | Install once, earn automatically |
| No wear | Only uses idle cycles, not 24/7 load |
| Flexible | Contribute when you want, pause when you need |
For Consumers
| Benefit | Description |
|---|---|
| Massive scale | Access thousands of devices instantly |
| Cost effective | Pay less than traditional cloud |
| No lock-in | No contracts, no minimums |
| Global reach | Compute nodes on every continent |
Use Cases
- 3D Rendering — Blender, animation studios, architectural visualization
- Scientific Research — Climate modeling, protein folding, simulations
- Machine Learning — Model training, hyperparameter tuning
- Video Processing — Transcoding, effects, batch processing
- Data Analysis — Large dataset processing, ETL pipelines
Join the Beta
OmnibusCloud is launching soon with distributed Blender rendering. Be among the first to:
- Contribute — Start earning with your idle devices
- Consume — Render your Blender projects at scale
Visit omnibuscloud.com to join the waitlist, or contact us to discuss enterprise needs.