by AITECH Cloud Network Foundation
AITECH Cloud Network pricing, plans and limits
Decentralized, hourly-billed NVIDIA GPU rental with no subscriptions or lock-in.
- Best for
- ML engineers running short-term fine-tuning or inference jobs
- Not for
- Compliance officers at regulated healthcare or finance companies needing an audited controls report
Should you use AITECH Cloud Network?
Best for developers and startups who need short bursts of GPU capacity without a long-term contract: choose from eight different NVIDIA GPU tiers and pay only for the hours used. AITECH is a young, decentralized marketplace, so it suits spot workloads better than mission-critical production traffic requiring published compliance certifications.
Pairing hourly, no-lock-in GPU rental with a built-in agent orchestration layer (Agent Forge) under one account.
Compute marketplaces are commoditized on price; AITECH's edge is bundling hourly GPU rental with its own Agent Forge orchestration product under one account, but decentralized peers like io.net and Akash Network compete on the same GPU-hour pricing model with larger published device pools.
Also for
- Indie AI developers testing models without a hyperscaler contract
- Indie AI developers and startups testing model training without committing to a hyperscaler account
Show 2 more
- Agent developers who want compute and orchestration (Agent Forge) from a single vendor
- GPU owners looking to list idle hardware capacity as a marketplace supplier
Also not for
- Enterprise buyers requiring a fixed, disclosed data-center region for data residency
Main trade-off
No published compliance certification (SOC 2, ISO 27001, HIPAA) and no fixed, disclosed data-center region list, which makes it hard for enterprise buyers to evaluate against centralized clouds.
AITECH Cloud Network pricing
Server plans
| Plan | GPU | Price |
|---|---|---|
| NVIDIA L40SGPU · Inference, rendering, and scalable deployment environments · global | 1× NVIDIA L40S, 48 GB VRAM | $0.80/hr |
| NVIDIA H100GPU · High-throughput AI training and advanced model development · global | 1× NVIDIA H100, 80 GB VRAM | $2.40/hr |
| NVIDIA H200GPU · Memory-intensive AI workloads and complex multi-step AI systems · global | 1× NVIDIA H200, 141 GB VRAM | $2.84/hr |
| NVIDIA B100GPU · global | 1× NVIDIA B100 | — |
| NVIDIA B200GPU · global | 1× NVIDIA B200 | — |
| NVIDIA B300GPU · global | 1× NVIDIA B300 | — |
| NVIDIA RTX 6000 AdaGPU · global | 1× NVIDIA RTX 6000 Ada | — |
| NVIDIA RTX 6000 ProGPU · global | 1× NVIDIA RTX 6000 Pro | — |
- Billing
- Hourly · Pay as you go
How the pricing works, in full
NVIDIA L40S starts at $0.80/hr per card, H100 at $2.40/hr, and H200 at $2.84/hr; AITECH states these are introductory rates for the initial release and may change. B100, B200, B300, RTX 6000 Ada, and RTX 6000 Pro instances, plus dedicated bare-metal clusters, are quote-only via AITECH's sales team, not published rates. No subscription or long-term contract is required; funding is by crypto or fiat credit deposit rather than card billing.
Costs that are easy to miss
- Pricing for the higher-end GPU tiers and bare-metal clusters is quote-only, so the real monthly cost for anything beyond the three headline tiers is unknown until a buyer contacts sales.
- Funding an account requires a crypto or fiat credit deposit rather than direct card billing, adding a treasury or operations step most centralized clouds skip.
AITECH Cloud Network regions and compliance
- Uptime SLA
- 99.9%
Available regions
- global
Access methods
- Web console
- Dashboard
What AITECH Cloud Network does
In April 2026, Solidus AI Tech rebranded as AITECH Cloud Network (ACN), a decentralized GPU cloud marketplace offering on-demand NVIDIA data-center GPUs for AI training, inference, and rendering across a distributed provider network rather than one company's own data centers.
Hourly Per-Card GPU Pricing
GPU instances are billed by the hour per card with no subscription or minimum term, so a workload can be spun up and torn down the same day.
Agent Forge Orchestration
A separate no-code layer for designing, deploying and managing autonomous AI agents, sold and used independently of the Compute Marketplace.
Dedicated Bare-Metal on Request
Enterprises needing isolated hardware can request bare-metal GPU clusters configured to specific compliance and performance requirements, though pricing is quote-only.
CertiK-Audited Smart Contracts
The $ACN token's smart contracts carry a CertiK audit, the only third-party security review AITECH publishes as of September 2026.
Supplier Onboarding
GPU owners can list spare capacity on the marketplace as a vendor, which is how AITECH sources supply without owning every data center itself.
Strengths and limits
Strengths
- Billing is purely hourly with no subscription, contract, or minimum term, so idle capacity costs nothing once an instance is stopped.
- Bundles raw GPU compute (Compute Marketplace) and agent orchestration (Agent Forge) under one account, useful for teams that want both without integrating two separate vendors.
- Supports both crypto and fiat funding, letting Web3-native teams and traditional companies use the same marketplace without a separate off-ramp.
Limits
- No published SOC 2, ISO 27001, HIPAA, or similar compliance certification, which rules it out for regulated workloads that require an audited controls report.
- Pricing beyond the three headline GPU tiers is quote-only ('Enquire Now'), so buyers cannot compare full costs upfront for the rest of the catalog or for dedicated bare-metal clusters.
- Funding a deployment involves depositing crypto or fiat credit to an account first, an extra step that centralized clouds with direct card billing do not require.
AITECH Cloud Network alternatives
| Service | Regions | Pricing | Compliance | Checked |
|---|---|---|---|---|
| AITECH Cloud Network This page | Global | Hourly · Pay as you go | -- | 28 Sep |
 AtlasCloudPick AtlasCloud if you need a single API across 300+ third-party models alongside GPU rental; pick AITECH if you only need raw hourly GPU compute plus its own Agent Forge orchestration. | Europe · Asia Pacific · US | Pay as you go · Hourly | SOC 2 · HIPAA | 23 Oct |
 BasetenPick Baseten if you need SOC 2 Type II or HIPAA-eligible infrastructure; pick AITECH if compliance certification is not a requirement and you want decentralized, no-lock-in pricing instead. | US · Europe · Asia Pacific | Free tier · Pay as you go · Hourly | SOC 2 · HIPAA · GDPR | 29 Oct |
 Crusoe CloudPick Crusoe Cloud if you need disclosed US/Norway data-center regions and free network egress; pick AITECH if you want a decentralized marketplace with no fixed region commitment. | US · Europe · Canada | Pay as you go · Hourly | SOC 2 | 19 Oct |
AITECH Cloud Network facts
- Open source
- No
What changed
Questions about AITECH Cloud Network
What does AITECH Cloud Network actually cost?
AITECH's Compute Marketplace bills by the hour per GPU card with no subscription: NVIDIA H100 starts at $2.40/hr, H200 at $2.84/hr, and L40S at $0.80/hr, per AITECH's own pricing page as of September 2026. The remaining higher-end tiers, plus dedicated bare-metal clusters, are quote-only through AITECH's sales team rather than self-serve checkout. AITECH states these introductory rates are for the initial release and subject to change.
Can you use AITECH Cloud Network without paying?
No, AITECH Cloud Network does not offer a free tier or trial credits. You fund an account with a crypto or fiat credit deposit before deploying any instance, and every GPU-hour is billed from the moment the instance launches, with no minimum spend or long-term contract required.
What are AITECH Cloud Network's closest competitors?
io.net, Akash Network, and Render Network are the best-known decentralized GPU marketplaces AITECH competes with. io.net markets the largest device supply and the lowest spot prices, Akash covers a broader set of cloud services beyond GPUs, and Render focuses on creative rendering workloads rather than general AI training.
How does AITECH Cloud Network compare to io.net in 2026?
Both are decentralized, hourly-billed GPU marketplaces, but io.net advertises a substantially larger aggregated device pool and the lowest published spot pricing, while AITECH pairs its Compute Marketplace with its own Agent Forge orchestration product under one account. Pick io.net for raw price and supply depth; pick AITECH if you also want built-in agent orchestration without adding a second vendor.
How do you set up AITECH Cloud Network?
Create an account on the Compute Marketplace, deposit crypto or fiat credit, then launch a GPU instance directly from the dashboard, a flow AITECH says takes minutes. Enterprises needing bare-metal or higher-end hardware go through AITECH's sales team ('Enquire Now') instead of self-serve checkout.
About AITECH Cloud Network
AITECH Cloud Network (ACN) is a decentralized GPU compute marketplace operated by AITECH Cloud Network Foundation, with the Compute Marketplace and Agent Forge orchestration platform run day-to-day by an affiliated entity, Solidus AI Tech Digital. The project began as Solidus AI Tech and rebranded to AITECH Cloud Network on April 22, 2026, alongside a token migration from BNB Chain to Ethereum. Rather than owning its own data centers outright, ACN aggregates GPU supply from multiple providers and lists it on a market-priced marketplace, competing with other decentralized compute networks such as io.net, Akash Network, and Render Network.
Read the full background
The core product is the Compute Marketplace: on-demand access to eight distinct NVIDIA GPU tiers, spanning workstation-class rendering cards up to flagship data-center training accelerators, billed hourly per card with no subscription tier or minimum term. AITECH also runs Agent Forge, a separate no-code orchestration layer for deploying and managing autonomous AI agents; both products can be used independently of each other and of the $ACN token. Dedicated bare-metal GPU clusters are available on request for buyers with specific compliance or performance requirements, and pricing above the three headline GPU tiers is quote-only through AITECH's sales team.
The marketplace targets AI companies, ML teams, developers, and research institutions that need short bursts of GPU capacity, such as fine-tuning a model or running a batch inference job, without committing to a reserved-instance contract from a hyperscaler. Funding an account requires depositing crypto or fiat credit ahead of time, a workflow unfamiliar to buyers used to direct card billing and a monthly invoice.
AITECH's smart contracts have been audited by CertiK, and $ACN token holders can stake through a separate staking product; that token economy is a distinct layer from the GPU rental business and is not required to use the Compute Marketplace. The company ships weekly platform updates: a recent development note described stability and runtime work on Agent Forge alongside a public discussion with identity-verification partner Concordium about safer autonomous-agent workflows.
For buyers comparing GPU cloud options on Hokai, AITECH sits in the GPU compute category alongside peers like AtlasCloud, Baseten, and Crusoe Cloud. As a young, decentralized marketplace it is best treated as a spot-priced supplement to a primary cloud provider rather than a sole source of production capacity, since it does not yet publish SOC 2, ISO 27001, or similar compliance certifications.
Sources and guides
Official sources
Plans, regions, the uptime commitment and the compliance list on this page are read from AITECH Cloud Network Foundation's own pages and documentation. The verdict is Hokai's reading of them, not the provider's claim.
Lists that include AITECH Cloud Network
More AI Infrastructure Services on Hokai
AgentMailEmail inbox API built so every AI agent can own, read and answer its own inbox
Amazon SESAWS email sending and receiving, metered per thousand messagesCloudflare Email ServiceSend and receive email from Workers and agents, with unlimited inbound routing
PostmarkTransactional email API that keeps app mail apart from bulk sendsResendEmail API for developers, with inboxes your AI agents can read and answer
SendGridTwilio email API and SMTP relay for high-volume transactional and marketing mail