A fast, unified way for businesses to generate API keys and connect to our global network of AI models and agents.
DeCenter AI is a unified AI studio that connects users to thousands of premium AI tools through a Super Agent that intelligently matches every prompt to the most effective model — for just $0.01 per inference.**.
The platform provides developers with easy API access to an AI network of 400+ models and agents, powered by Hedera’s Consensus Service (HCS) for transparent inference tracking and Hedera Token Service (HTS) for seamless microtransactions in HBAR. DeCenter AI bridges AI & DePIN — combining smart autonomous AI systems with decentralized infrastructure for trust, scalability, and accessibility.**.
Today’s AI landscape is fragmented and expensive.
- Too Many Options: Hundreds of apps that do the same thing with minor variations, creating confusion and inefficiency.
- Too Many Payments: Every tool requires a separate subscription, making AI usage costly and unsustainable.
- Too Much Hassle: Users switch between multiple platforms just to complete a single task, wasting time and energy.
We are building a Unified AI Studio that connects users to thousands of premium AI tools, powered by a Super Agent that automatically selects the best-performing model for every prompt — for just one cent per inference.
For the hackathon, we built a B2B platform that allows developers and organizations to:
- Generate API keys to access our AI network.
- Connect directly to over 400+ AI models and agents.
- Pay just $0.01 per inference.
This enables anyone to power their applications with high-quality AI — seamlessly and affordably.
Our architecture is built on three core principles:
- Agentic: Intelligent orchestration of AI agent results.
- Agnostic: Accessible across any connected device or platform.
- Abstracted: API-native for effortless integration and deployment.
- LangGraph / LangChain – Dynamic AI orchestration.
- ILM (Inference Lifecycle Management) – Performance optimization.
- Hedera Consensus Service (HCS) – Records inference logs and verification hashes.
- Hedera Token Service (HTS) – Processes microtransactions in HBAR via testnet wallets.
Workflow:
Prompt → AI Selection (via Super Agent) → Hedera Registry (HCS Logging) → Output Delivery
DeCenter AI empowers developers and businesses to:
- Access premium AI capabilities without enterprise pricing barriers.
- Build trustworthy AI applications with verifiable inference records.
- Reduce AI infrastructure costs by over 90% using decentralized micro-billing.
- Support community-powered AI infrastructure running on Hedera, aligning with the AI & DePIN mission of building smarter, decentralized systems.
+------------------------------+
| User Interface (Prompt Input)|
+--------------+---------------+
|
v
+--------------+--------------+
| Super Agent (Model Selector) |
+--------------+--------------+
|
v
+--------------+--------------+
| Model / Agent Inference |
+--------------+--------------+
|
v
+--------------+--------------+
| Hedera Registry (HCS) Log |
+--------------+--------------+
|
v
+--------------+--------------+
| Hedera Token Service (HTS) |
| - Microtransactions |
| - Reward Distribution |
+--------------+--------------+
|
v
+--------------+--------------+
| Result Verification (HCS) |
+--------------+--------------+
|
v
+------------------------------+
| Final Output Delivery to User|
+------------------------------+
System Flow
Prompt
→ Super Agent (Model Selection via LangChain/LangGraph)
→ Inference Lifecycle Management (ILM)
→ Hedera Consensus Service (HCS Logging)
→ Output Delivery
Tech Stack
- Frontend: React / Next.js
- Backend: Node.js
- AI Layer: LangChain, LangGraph, OpenAI, etc.
- Blockchain Layer: Hedera HCS + HTS for transparency and payments
- Database: Supabase
- Transparency: HCS ensures every AI inference is verifiable and tamper-proof.
- Microtransactions: HTS enables pay-per-inference pricing in HBAR.
- Efficiency: Hedera’s low-cost, carbon-negative network supports affordable large-scale AI operations.
- Trust Layer: Hedera provides decentralized auditability for AI decision-making and logging.
Yes. DeCenter AI started with a clear and real problem: AI infrastructure is fragmented, expensive, and difficult to scale for consumer applications. The initial concept of a unified AI studio with agent-based routing was well-defined and aligned with real market needs.
The baseline integration with Hedera was strong and intentional.
- Hedera Consensus Service (HCS) was used for logging AI inference activity
- Hedera Token Service (HTS) enabled microtransactions for pay-per-inference This created a clear alignment between AI infrastructure and decentralized trust/logging.
The project was technically feasible from the start.
- Modular architecture (API + AI routing + blockchain logging)
- Use of existing frameworks (LangChain, Supabase, Node.js)
- Clear execution path for MVP delivery within hackathon timeline
The product vision was clear: A single API to access multiple AI models at low cost, with transparent logging and micro-billing. The value proposition (reduce cost, simplify access, improve scalability) was easy to understand and compelling.
Significant improvements were made:
- Expanded from core API infrastructure to include Explorer (distribution layer)
- Improved developer experience with API key generation and playground testing
- Introduced built-in billing abstraction for end users
- Enhanced UX for both developers (B2B) and consumers (discovery layer)
This evolved the product from a tool into a full ecosystem (infra + distribution).
Technical depth increased meaningfully:
- Improved AI orchestration using agent-based routing (LangChain / LangGraph)
- Better handling of inference lifecycle management
- Optimized system flow between off-chain AI processing and on-chain logging
- Strengthened backend architecture with scalable API layer and Supabase integration
Strong post-baseline validation:
- Grew to 36+ B2B startups onboarded within 2 months
- Achieved 5M+ monthly inferences
- Secured ~$1M in infrastructure funding, credits, and grants
- Completed accelerator cohort (Foundance)
- Built early investor relationships and pipeline
This demonstrates real market demand and early product-market fit signals.
Deeper and more meaningful Hedera usage:
- Expanded use of HCS for verifiable inference logs and auditability
- Strengthened HTS microtransaction flow for scalable billing
- Improved architecture for real-time logging + retrieval via mirror nodes
- Positioned Hedera as a core trust and transaction layer in AI infrastructure
Hedera evolved from a supporting component to a critical backbone of the system.
-
Completed accelerator program (Foundance)
-
Established key partnerships with:
- Intel
- Google Labs
- Cloudflare
- Perplexity
-
Developed tokenomics model for network incentives
-
Built strong investor pipeline and fundraising momentum
-
Expanded vision to include AI + DePIN ecosystem infrastructure
Then follow these steps to create and use your API key:
- Sign in with google on https://decen-hedera.vercel.app/
- Go to the API page.
- Click Generate API, then enter the requested details.
- Click Generate.
- Your API key will be created in seconds—copy and use it.
To test the inferencing power of the DeCenter AI Network:
- Go to the Playground page.
- Type in a prompt.
- Hit Send.
- Watch the AI model/agent respond in real time.
- Monitor inferences logged on Hedera Testnet.
Stay up-to-date with the latest developments and news from DeCenter AI:
DeCenter AI bridges AI and DePIN by combining decentralized infrastructure with intelligent AI routing. Our goal is to make AI accessible, affordable, and verifiable for the 97% of users currently priced out of advanced AI systems.
