Your monthly cloud signal is live.

From livestream AMAs to infrastructure launches, over the last month, ASI:Cloud reinforced its position as the permissionless backbone for AI, Web3, and the Artificial Superintelligence Alliance.

We dropped a major new model, activated enterprise-grade GPU clusters in Sweden, and deepened real-world builder access through Imperial College and the ASI:Accelerator.

This is what’s inside:

• 🎙️ AMA With CUDOS Leadership: Building the AI Backbone for Web3
• 🏫 Imperial College Hackathon & Keynote: Showcasing ASI:Cloud for permissionless AI
• 🌍 Sweden GPU Cluster Goes Live: Enterprise-scale infrastructure from Singularity Compute, Conapto, and CUDO Compute
• 🖥️ Website Refresh: New UX, faster flows, and more bonuses for using $FET
• 🧠 New Inference Model Live: Production-ready open models added to the API
• 🚀 ASI:Accelerator Accepting Applicants: GPU access, mentorship, and go-to-market support
• 🎁 $20 Free Inference Credits: Deploy top models instantly with no KYC

🎙️ Building the AI Backbone for Web3 AMA with Luke & Matt

Matt Hawkins (CEO & Cofounder, CUDOS) and Luke Gniwecki (Head of Blockchain & AI Compute) went live on November 27th to share how ASI:Cloud is shaping the AI compute backbone for the ASI Alliance and Web3.

Topics covered:

• Real trends from GPU deployments
• ASI:Cloud powering smart contract AI, validator workloads, and inference
• Wallet-first infra vs legacy cloud

📺 Watch the complete AMA & discussion on X

🏫 Imperial College Hackathon & Keynote

We brought ASI:Cloud to the heart of innovation at Imperial College London. Alongside Fetch.ai and SingularityNET, the CUDOS team showcased how wallet-based access to GPU infrastructure is unlocking new possibilities for builders, researchers, and decentralized teams.

Luke Gniwecki (Head of Blockchain & AI Compute, CUDOS) delivered a keynote exploring how inference is shifting from a manual process to a programmable, service-based layer for developers.

🗣️ Key themes:

• AI tokens as programmable units of intelligence
• The shift from training to real-time inference
• Low-latency, wallet-first infrastructure for AI deployment
• ASI:Cloud as the foundation for scalable, open compute in AI + Web3

Rebekah Pennington (Marketing Manager, CUDOS) introduced students and researchers to the ASI:Accelerator, focusing on how the program empowers AI-native teams to go from idea to inference.

💡 What was covered:

• GPU infrastructure + mentorship for startups and researchers
• Go-to-market support aligned with the ASI Alliance
• Inference credits and prizes for top hackathon projects
• Real-world support from idea validation to production deployment

Thanks to Imperial College, Imperial AI Group, the Imperial Blockchain & FinTech society, and everyone who participated in the event. ASI:Cloud is giving the next generation of builders real access to scalable wallet-based AI infrastructure

🌍 GPU Cluster Launch in Sweden

Singularity Compute, in partnership with CUDOS and Conapto, activated an enterprise-grade GPU cluster in Sweden to power ASI:Cloud.

What’s live now:

• Bare-metal and VM GPU rentals
• Serverless + dedicated inference endpoints
• Enterprise-ready R&D and fine-tuning support

📖 Read the full press release here

🔍 Learn more about the infrastructure vision

🖥️ Website Refresh

The ASI:Cloud platform got a complete upgrade. We’ve rolled out a streamlined interface with faster load times, simplified navigation, and clearer model filtering.

New features include:

• Instant API key generation for quick starts
• Token usage tracking
• Expanded model details with performance tips and use cases
• Tiered deposit bonuses up to $12K as you scale
• Save up to 5% when staking $FET
• Get 5%+ more bonus credits when you pay with $FET

Whether you’re experimenting with ASI-1 Mini or scaling enterprise workloads, the new user experience makes it easier than ever to deploy permissionless AI.

Explore the refreshed layout and platform

🧠 We Added Another Inference Model

We’ve expanded the ASI:Cloud API with a powerful new inference model, purpose-built for real-world AI applications from reasoning agents to multilingual chat and coding.

Hermes-4–70B: A hybrid reasoning model fine-tuned on LLaMA 3.1–70B by Nous Research. Ideal for assistant-style use cases, long-context tasks, and structured logic across math, code, and creative writing.

Hermes drops directly into ASI:Cloud’s permissionless stack, accessible instantly through wallet-based login and a fully OpenAI-compatible API.

🧠 Deploy and test Hermes-4–70B now

We will have more new models available in 2026.

🚀 ASI:Accelerator Accepting Startups

We’re actively onboarding new AI-native teams to the ASI:Accelerator, a program designed to help developers, researchers, and Web3 startups go from idea to inference with full support from the ASI Alliance.

🌐 What’s included:

• Wallet-based access to GPU infrastructure on ASI:Cloud
• Mentorship across AI, blockchain, and go-to-market strategy
• Grants and ASI-funded inference credits to support early development
• Direct ecosystem connections with SingularityNET, Fetch.ai, and other Alliance partners

Whether you’re building LLM-powered agents, DePIN infrastructure, or research-backed AI apps, the ASI:Accelerator gives you the infra, guidance, and funding to scale fast.

🔗 Apply for ASI:Accelerator program now

🎁 Claim $20 in Free Inference Credits

Want to test ASI:Cloud?

We’re giving out $20 in free inference credits so you can try to open models like LLaMA 3.3, Qwen3–32B, Hermes-4–70B, and ASI-1 Mini, which is always free to use.

✅ What you get access to:

• Serverless inference with pay-per-token pricing
• One-click deployment via templates
• Free and unlimited use of ASI-1 Mini
• Wallet-based access with OpenAI-compatible APIs

📥 Redeem in 3 easy steps:

• 1️⃣ Sign up to ASI:Cloud
• 2️⃣ Join our ASI:Cloud Discord Community
• 3️⃣ Fill out the inference credit form

💡 What’s Next for ASI:Cloud

Everything coming soon on the ASI:Cloud roadmap:

• 💸 Fiat payments, credit card payments, and recurring billing
• 🤖 More inference models with two more available soon
• 🗣️ Speech-to-text and voice model integration
• 🖼️ Image + video generation (multi-modal)
• 🧠 Knowledge Graph pipelines
• 🏢 Enterprise endpoints and private clusters
• 📱 Mobile-accessible deployment templates

Built for inference, backed by NVIDIA GPUs, powered by CUDOS, and co-developed by SingularityNET, ASI:Cloud is the compute layer of the ASI Alliance.

Stay up to date on everything on ASI:Cloud by signing up for our newsletter, joining our conversations on X & Linkedin, and being part of our Telegram & Discord communities.

Download Complete.

The ASI:Cloud Download — December 2025 was originally published in CUDOS on Medium, where people are continuing the conversation by highlighting and responding to this story.

Your monthly cloud signal is live.

From live panel appearances to massive milestones, November saw ASI:Cloud continue to accelerate as the inference layer for permissionless AI.

We crossed 6 billion tokens processed, scaled to 1,000+ active inference users, launched the ASI:Accelerator, and supported the ASI Alliance’s effort alongside CUDO Compute to provide decentralized AI infrastructure for Cocoon AI on Telegram.

This is what’s inside:

• 📈 6 Billion Tokens Processed: Milestone unlocked for serverless inference at scale
• 👥 1,000+ Active Inference Users: Startups, builders, and researchers deploying live
• 🎙️ Mansion House Summit Series Panel Recap: Luke Gniwecki spoke on permissionless AI at Mansion House, London
• 🎓 Imperial College Keynote: Supercharging AI Development with ASI:Cloud
• 🚀 ASI:Accelerator Launch: Mentorship, infra, and GPU access for AI-native teams
• 🧠 AiMo Joins Accelerator: Building the agent-native economy on ASI:Cloud
• 🤝 Scaling Cocoon AI on Telegram: Supporting privacy-first AI with distributed compute
• 🎁 $20 Free Inference Credits: Deploy open-source LLMs instantly

6 Billion Inference Tokens Processed

ASI:Cloud has processed over 6 billion inference tokens since launch.

That’s enough compute to:

• Document every major GitHub repo
• Caption every movie + TV episode ever made (twice)
• Read the entire English Wikipedia… 20 times

And it all happened without signups, accounts, or vendor lock-in. Just wallet-based, pay-per-token access on a permissionless stack.

👉 Start deploying on ASI:Cloud

1,000+ Active Inference Users

Inference is live and growing: over 1,000 active builders are running models through ASI:Cloud’s serverless API layer. Startups, researchers, protocol teams, and enterprises are all deploying on demand.

That’s all achieved with:

• ✅ No KYC
• ✅ Wallet-first access
• ✅ Open models, OpenAI-compatible APIs
• ✅ Pay-per-token pricing on NVIDIA GPUs

🧠 Try ASI-1 Mini free

Keynote at Imperial College London

Luke Gniwecki delivered a live lecture on “Supercharging AI Development with ASI:Cloud” at Imperial College as part of the Intro to AI Lecture Series.

Covered topics included:

• Open vs closed LLM architectures
• Wallet-based GPU access
• Inference deployment without vendor lock-in
• Scaling real-world AI apps via ASI:Cloud

CUDOS on Stage at GDASW3

At GDASW3 London, Luke Gniwecki (CUDOS), Maria Minaricova (Fetch.ai), and David Taylor (W3CMO) explored how ethical AI systems must be backed by open infrastructure.

Key points:

• Infrastructure determines intelligence
• Open, sovereign systems > walled gardens
• Permissionless inference = AI alignment at scale

Thank you to The Digital Commonwealth and partners for an incredible event.

📸 See some of the event moments

ASI:Accelerator Program Launch

We officially launched the ASI:Accelerator program.

Aimed at AI-native startups and builders, the Accelerator delivers:

• Wallet-based access to ASI:Cloud
• GPU-backed compute with no KYC
• Mentorship, infrastructure credits, and go-to-market support
• Ecosystem alignment with the ASI Alliance

🔗 Apply now

AiMo Network: First Accelerator Participant

CUDOS officially welcomes AiMo Network to the ASI:Accelerator.

AiMo is building a permissionless, agent-native AI economy that connects humans, agents, and services through on-chain inference.

ASI:Cloud will power their deployments with:

• Crypto-native inference
• Wallet-based GPU rentals
• OpenAI-compatible APIs
• No lock-ins, no borders

📖 Read the press release

🔗 Explore AiMo Network

Scaling Cocoon AI with Distributed Compute

ASI Alliance, CUDO Compute, AlphaTON Capital, Vertical Data, and SingularityNET have partnered to accelerate Cocoon AI on Telegram.

Privacy-first and blockchain-aligned, Cocoon AI is one of the largest decentralized AI deployments to date, backed by:

• 🔐 User-owned, privacy-preserving AI

• 🌍 Renewable-powered GPU infra

• 🧠 On-chain alignment through ASI:Cloud

📣 Read the full announcement

Claim $20 in Free Inference Credits

Get started on ASI:Cloud with $20 in free inference credits. Ideal for testing open-source LLMs, running your first deployments, or exploring the platform. No KYC, no credit card, just fast access to serverless AI on NVIDIA-backed infrastructure.

Claim in 3 simple steps:
1️⃣ Sign up to ASI:Cloud

2️⃣ Join our ASI:Cloud Discord Community

3️⃣ Fill out the inference credit form

What’s Next for ASI:Cloud

Coming soon to the ASI:Cloud platform:

• 🗣️ Voice models for real-time speech-to-text and audio processing

• 🖼️ Multi-modal support, including image and video generation

• 🧠 Knowledge Graph creation to power structured, linked intelligence

• 🏢 Enterprise-grade scaling with dedicated endpoints and GPU clusters

• 💳 Fiat and credit card payments for broader team access

• 🔄 Expanded model catalog with more open-source LLMs

Built for inference, backed by NVIDIA GPUs, powered by CUDOS, and co-developed by SingularityNET, ASI:Cloud is the compute layer of the ASI Alliance.

Stay up to date on everything on ASI:Cloud by signing up for our newsletter, joining our conversations on X & Linkedin, and being part of our Telegram & Discord communities.

Download Complete.

The ASI:Cloud Download — November 2025 was originally published in CUDOS on Medium, where people are continuing the conversation by highlighting and responding to this story.

Accelerating On-Chain Intelligence with Permissionless Inference

We’re excited to announce AiMo Network as the first official participant in the ASI:Accelerator Program in collaboration with Artificial Superintelligence Alliance, a key milestone in our mission to support builders creating the next generation of permissionless AI infrastructure.

AiMo is building the Intelligence Capital Market On-Chain: a new layer where humans, agents, and service providers interact via a permissionless API, agent-native interface, and pay-per-inference model powered by x402.

By integrating with ASI:Cloud as one of its infrastructure partners, AiMo gains access to wallet-based GPU compute and high-performance inference infrastructure without centralized gatekeepers, vendor lock-in, or KYC.

“AiMo embodies exactly what the ASI:Accelerator is designed for visionary teams building open AI ecosystems using distributed compute and crypto-native access.”
– Luke Gniwecki, Head of Blockchain & AI Compute Product at CUDOS

This collaboration brings together shared values:

• Open, censorship-resistant design
• Permissionless APIs and wallet-first onboarding
• Scalable, serverless inference on enterprise-grade infrastructure
• Alignment with the broader mission of the ASI Alliance

As the first Accelerator participant, AiMo will work closely with the ASI:Cloud team to:

• Optimize AI workloads on distributed GPU infrastructure
• Leverage OpenAI-compatible APIs with pay-per-token access
• Enable frictionless deployment of agent-native application

About ASI:Accelerator

The ASI:Accelerator is a program created to support early-stage AI, agent, and DePIN teams building in the open AI ecosystem. Participants receive infrastructure credits, technical integration support, and access to ASI:Cloud’s wallet-based GPU compute and serverless inference tools.

About CUDOS

CUDOS operates the distributed infrastructure powering ASI:Cloud. With over two decades of expertise in managing large-scale compute, CUDOS delivers enterprise-grade performance, reliability, and scalability for AI, Web3, and blockchain applications. As a core member of the ASI Alliance, CUDOS is building the decentralized backbone for permissionless, AI-native infrastructure.

About AiMo Network

AiMo Network is building the Intelligence Capital Market On-Chain. Through a permissionless API, pay-per-inference model (x402), and agent-native interface, AiMo connects humans, AI agents, and services without borders, censorship, or centralized control. The network enables a new era of open, interoperable intelligence infrastructure.

Learn More

🔗 Visit AiMo Network
🔗 Explore the ASI:Accelerator
🔗 Start deploying on ASI:Cloud

Powered by CUDOS, co-developed by SingularityNET, and built in collaboration with the ASI Alliance.

AiMo Network Joins ASI:Accelerator was originally published in CUDOS on Medium, where people are continuing the conversation by highlighting and responding to this story.