ASI:Cloud Exits Beta And Starts Delivering Enterprise-Grade AI Workloads
London, United Kingdom, December 17th, 2025, Chainwire
Artificial Super Intelligence Alliance has announced that ASI:Cloud is now processing live workloads, offering developers permissionless access to enterprise-grade GPU infrastructure and AI inference services.
Rather than forcing AI builders to navigate fragmented vendor ecosystems, the ASI:Cloud platform provides unified infrastructure. This is particularly valuable given that most enterprises intend to repatriate their workloads, sending institutional demand for production AI soaring. ASI:Cloud provides teams with a robust and flexible alternative to centralized vendors.
Ben Goertzel, CEO of SingularityNET and the ASI Alliance, said: “Every GPU cycle brings us closer to sustainable decentralized intelligence. ASI:Cloud represents that vision becoming operational infrastructure rather than aspiration.”
ASI:Cloud combines permissionless infrastructure with transparent pricing, allowing developers to authenticate using Web3 wallets without KYC. Payments can be made in FET tokens and stablecoins, with fiat options coming soon.
The platform delivers OpenAI-compatible inference endpoints across top open-source models, including Llama 3.3 70B, Qwen 3 32B, and Gemma 3 27B, at competitive pricing starting at $0.07 per million input tokens.
In comparison, operational instances of the popular H100 GPU on major cloud platforms such as AWS and Azure typically range from $3.90 to $6.98 per hour. At the same time, bandwidth, egress, and storage incur unpredictable surcharges, limiting accessibility.
Built in collaboration between SingularityNET – which supplies the AI infrastructure backend and model optimization – and CUDOS – which operates enterprise-grade compute infrastructure globally – ASI:Cloud unifies billing, developer experience, and network access.
Luke Gniwecki, Head of AI Compute Product at CUDOS said: “Enterprises are hitting real constraints in the centralized cloud market, with capacity shortages and vendor lock-in holding back AI innovation at precisely the moment demand is exploding. ASI:Cloud combines CUDOS’s enterprise-grade compute infrastructure with SingularityNET’s AI backend. This gives builders dependable access to high-performance compute without the bottlenecks that define the legacy providers.”
The infrastructure arrives at a critical inflection point as Nvidia’s latest GPU allocations are sold out through 2026, with waitlists extending well into next quarter, and the decentralized computing market projected to reach $45 billion by 2035.
Developers and researchers interested in exploring the platform can head over to asicloud.cudos.org. Documentation is available at docs.cudos.org.
About Artificial Superintelligence Alliance
The Artificial Super Intelligence (ASI) Alliance is a collective formed by Fetch.ai, SingularityNET, and CUDOS. As the largest open-sourced, independent entity in decentralized AI research and development, the alliance aims to accelerate advancement of decentralized Artificial General Intelligence (AGI) and, ultimately, Artificial Superintelligence (ASI).