Theta EdgeCloud
Hybrid cloud-edge GPU marketplace with 50-70% cost savings
Theta EdgeCloud is a hybrid cloud-edge computing platform that combines 30,000+ distributed edge nodes with Google Cloud and AWS partnerships, offering GPU instances for AI training, inference, video transcoding, and rendering at 50-70% cost savings versus traditional hyperscalers.
Available GPUs
Hourly on-demand pricing. Click column headers to sort.
Prices last updated: April 16, 2026
GPU Model↑ | Memory↑ | GPUs | vCPUs | RAM | Price / hr↑ | Updated↑ | Source |
|---|---|---|---|---|---|---|---|
| A100 SXM | 80GB | 1×2×4×8× | 10 | 80 GB | $1.99/hr | 4/15/2026 | |
| H100 SXM | 80GB | 1×2×4× | 10 | 80 GB | $2.29/hr | 4/15/2026 | |
| H200 | 141GB | 1×2×4× | 8 | 141 GB | $2.29/hr | 4/16/2026 | |
| Tesla T4 | 16GB | 1×2× | 2 | 8 GB | $0.20/hr | 4/16/2026 | |
| Tesla V100 | 32GB | 1×2× | 4 | 16 GB | $0.99/hr | 4/16/2026 |
Pros & Cons
Advantages
- 50-70% cost savings compared to AWS, Azure, and GCP
- Hybrid architecture combining cloud and edge for flexible scaling
- 30,000+ globally distributed edge nodes providing massive compute capacity
- Strong institutional customers including Stanford, KAIST, and NTU Singapore
- Backed by Samsung, Sony, Google, and Binance as enterprise validators
- No vendor lock-in with single API for containerized jobs
Limitations
- Dynamic marketplace pricing with no fixed published rates
- Relies on crypto token ecosystem (TFuel) for payments and rewards
- Community node reliability can vary compared to dedicated data centers
- GPU availability and pricing require authentication to view
- Limited to NVIDIA GPU offerings (H100, A100, 4090, 3090)
Key Features
Hybrid Cloud-Edge Architecture
Combines traditional cloud GPUs via Google Cloud and AWS with 30,000+ community-operated edge nodes for flexible compute.
Decentralized GPU Marketplace
Supply-demand driven pricing where node operators set rates and users select GPUs based on availability and cost.
Intelligent Workload Routing
Routes heavy training jobs to cloud/datacenter GPUs and distributes parallelizable inference across edge nodes.
Automatic Failover
Reroutes jobs to healthy nodes if a community node goes offline mid-task, ensuring workload continuity.
Containerized Workloads
Docker-based job execution with Jupyter Notebook and SSH access for flexible development environments.
AI Service APIs
On-demand model inference APIs and dedicated AI model serving with support for popular generative AI models.
Agentic AI Platform
RAG-powered AI agents with custom tools, live escalation, and integration capabilities for enterprise applications.
Persistent Storage
Persistent storage solutions for GPU nodes enabling stateful workloads and data persistence across sessions.
Compute Services
EdgeCloud GPU Instances
On-demand GPU instances from distributed edge nodes and cloud partnerships.
EdgeCloud AI Services
AI-specific services and APIs for model deployment and inference.
EdgeCloud Video Services
Specialized video processing and streaming services.
Pricing Options
| Option | Details |
|---|---|
| Marketplace Pricing | Dynamic pricing set by node operators in a supply-demand GPU marketplace. |
| On-Demand | Hourly billing with no long-term commitments or contracts required. |
Availability & Support
Regions
30,000+ globally distributed edge nodes with cloud capacity via Google Cloud and AWS regions.
Support
Documentation at docs.thetatoken.org, dashboard monitoring, and community support channels.
Getting Started
- 1
Create a Theta EdgeCloud account
Sign up at thetaedgecloud.com to access the GPU marketplace dashboard.
- 2
Browse available GPU instances
Explore available H100, A100, 4090, and 3090 instances with marketplace pricing from node operators.
- 3
Select GPU and configure workload
Choose a GPU instance and configure your containerized workload with Docker, Jupyter Notebook, or SSH access.
- 4
Deploy and monitor
Launch your workload with automatic failover and monitor progress through the dashboard.
Compare Providers
Find the best prices for the same GPUs from other providers
