Price comparisons of DePIN networks against AWS.
Hourly rates across networks
Sourced from public order books, marketplace leases, and on-demand cloud pricing.
| Network | GPU model | VRAM | Hourly rate | Billing model | Status | Access |
|---|---|---|---|---|---|---|
Render NetworkChain: Solana | NVIDIA GeForce RTX 4090 (24GB) | 24 GB GDDR6X | $0.38 / hrLast checked: 2026-09-28 | job based | Last checked: 2026-09-28 | Deploy |
io.netChain: Solana | NVIDIA GeForce RTX 4090 (24GB) | 24 GB GDDR6X | $0.42 / hrLast checked: 2026-09-30 | hourly | Last checked: 2026-09-30 | Deploy |
Akash NetworkChain: Cosmos | NVIDIA GeForce RTX 4090 (24GB) | 24 GB GDDR6X | $0.45 / hrLast checked: 2026-10-01 | dynamic bid | Last checked: 2026-10-01 | Deploy |
AethirChain: Arbitrum | NVIDIA GeForce RTX 4090 (24GB) | 24 GB GDDR6X | $0.48 / hrLast checked: 2026-09-29 | hourly | Last checked: 2026-09-29 | Deploy |
Akash NetworkChain: Cosmos | NVIDIA A100 (80GB SXM4) | 80 GB HBM2e | $1.15 / hrLast checked: 2026-10-01 | dynamic bid | Last checked: 2026-10-01 | Deploy |
io.netChain: Solana | NVIDIA A100 (80GB SXM4) | 80 GB HBM2e | $1.20 / hrLast checked: 2026-09-30 | hourly | Last checked: 2026-09-30 | Deploy |
AethirChain: Arbitrum | NVIDIA A100 (80GB SXM4) | 80 GB HBM2e | $1.35 / hrLast checked: 2026-09-29 | hourly | Last checked: 2026-09-29 | Deploy |
Akash NetworkChain: Cosmos | NVIDIA H100 (80GB HBM3) | 80 GB HBM3 | $1.85 / hrLast checked: 2026-10-01 | dynamic bid | Last checked: 2026-10-01 | Deploy |
io.netChain: Solana | NVIDIA H100 (80GB HBM3) | 80 GB HBM3 | $1.98 / hrLast checked: 2026-09-30 | hourly | Last checked: 2026-09-30 | Deploy |
AethirChain: Arbitrum | NVIDIA H100 (80GB HBM3) | 80 GB HBM3 | $2.10 / hrLast checked: 2026-09-29 | hourly | Last checked: 2026-09-29 | Deploy |
Amazon Web Services (AWS)BenchmarkHyperscaler | NVIDIA A100 (80GB SXM4) | 80 GB HBM2e | $3.06 / hrLast checked: 2026-10-02 | hourly | Last checked: 2026-10-02 | Deploy |
Amazon Web Services (AWS)BenchmarkHyperscaler | NVIDIA H100 (80GB HBM3) | 80 GB HBM3 | $3.88 / hrLast checked: 2026-10-02 | hourly | Last checked: 2026-10-02 | Deploy |
Render NetworkChain: Solana | NVIDIA H100 (80GB HBM3) | 80 GB HBM3 | Not verified yetLast checked: 2026-10-01 | job based | Not verified yet | Deploy |
Render NetworkChain: Solana | NVIDIA A100 (80GB SXM4) | 80 GB HBM2e | Not verified yetLast checked: 2026-10-01 | job based | Not verified yet | Deploy |
Amazon Web Services (AWS)BenchmarkHyperscaler | NVIDIA GeForce RTX 4090 (24GB) | 24 GB GDDR6X | Not verified yetLast checked: 2026-10-02 | hourly | Not verified yet | Deploy |
Profiles, consensus mechanisms, and hardware supply across leading networks.
Render Network is a leading decentralized GPU network founded by OTOY, originally designed for distributed 3D rendering and now expanded into generative AI inference, NeRFs, and spatial computing. It connects node operators with spare consumer and workstation GPU capacity to digital creators and ML engineers requiring scalable graphics and compute power.
io.net is a state-of-the-art DePIN network that aggregates underutilized GPUs from independent data centers, crypto mining facilities, and consumer nodes into unified, low-latency clusters. Built on the Ray distributed computing framework, io.net allows AI engineers to deploy massive GPU clusters for model training, fine-tuning, and scalable batch inference at up to 70–80% lower cost than traditional hyperscalers.
Akash Network is a sovereign Proof-of-Stake blockchain and permissionless compute marketplace built with the Cosmos SDK. It allows any cloud provider or data center to list idle CPU, GPU, memory, and storage capacity in an open reverse auction. Developers deploy containerized Docker applications using human-readable Stack Definition Language (SDL) files, achieving up to 85% savings over centralized cloud providers.
Amazon Web Services (AWS) is the world's leading centralized cloud platform, offering accelerated compute through Amazon EC2 instance families including P5 (H100 SXM5), P4de (A100 80GB), and G5 (A10G). It represents the industry benchmark against which decentralized alternatives compete, delivering unmatched networking, certifications, and compliance at premium enterprise pricing.
Aethir is an enterprise-grade decentralized physical infrastructure network (DePIN) specialized in high-performance GPU compute. Rather than relying solely on consumer hardware, Aethir aggregates enterprise-grade NVIDIA H100, A100, and RTX cards distributed across global tier-3 and tier-4 data centers, delivering ultra-low latency compute for enterprise AI pipelines and interactive cloud gaming.
Side-by-side architecture teardowns, pricing gaps, and workload recommendations.
Akash Network and Render Network serve complementary niches in the decentralized compute ecosystem. Akash Network stands out for fully open-source, permissionless, and censorship-resistant marketplace with zero vendor lock-in, whereas Render Network excels in deep integration with industry-standard graphics pipelines including octanerender, cinema4d, and blender.
io.net and Render Network serve complementary niches in the decentralized compute ecosystem. io.net stands out for io cloud clusters disparate geographic gpus using the ray framework with automated orchestration, whereas Render Network excels in deep integration with industry-standard graphics pipelines including octanerender, cinema4d, and blender.
Akash Network is the superior choice for AI startups, researchers, and cost-sensitive engineering teams who want to cut GPU compute bills by 50% to 75% without lengthy contract commitments. AWS remains the required standard for highly regulated Fortune 500 enterprises needing strict HIPAA/FedRAMP certification, ultra-large 10,000+ GPU InfiniBand clusters, and contractually insured financial 99.99% SLAs.
io.net is the superior choice for AI startups, researchers, and cost-sensitive engineering teams who want to cut GPU compute bills by 50% to 75% without lengthy contract commitments. AWS remains the required standard for highly regulated Fortune 500 enterprises needing strict HIPAA/FedRAMP certification, ultra-large 10,000+ GPU InfiniBand clusters, and contractually insured financial 99.99% SLAs.
Curated evaluations matching specific workloads with the optimal network and accelerator.
Low-latency, cost-effective serving for open-weight LLMs, vision, and speech models
High-throughput parallel compute for OctaneRender, Blender, Cinema4D, and Unreal Engine
Cost-effective training runs for LoRA, QLoRA, and domain adaptation under $2/hour per GPU
Reporting on GPU inventory, protocol upgrades, and cloud economics.
The median on-demand H100 price was $3.42 per GPU-hour across 40 providers in the week of September 28. CME plans to launch GPU rental futures on October 5, pending regulatory review.
Aethir says ACCELERATE gives access to 10 sites totaling up to 20 MW and projects up to $700 million in contracts by end of 2026. These are company projections, not completed contracts.
Salad published the implementation design for milestone 3 of its Render Network integration, which routes every transaction through the burn-mint-equilibrium model. Customer RENDER payments would go to the customer's own on-chain settlement account.
Disclosure: Some links on this page may be affiliate referral links. When you lease compute through these links, DecentralGPU may earn a small referral commission at no additional cost to you. See our affiliate disclosure and pricing methodology.