Haithe - Decentralized Protocol for Verifiable AI

jriyyya · June 12, 2025, 8:06pm

Haithe: A Decentralized AI Marketplace Ecosystem

Team Details

Spandan Barve - (@marsian83)
Riya Jain - (@jriyyya)
Kartikay Tiwari - (@ishtails)

Relevant Links

Try out Haithe here: https://haithe.hetairoi.xyz
GitHub Repository: https://github.com/Haithedotai/core
Watch Tutorial: https://youtu.be/2HCJYMXJGDk?si=KHq6MzZLolLXOHuA
Pitch Deck: Haithe - Pitch Deck
X Account: https://x.com/haithedotai
Haithe’s official bot: Telegram: Launch @haithebot
Telegram Community: Telegram: View @haitheai

The Vision

Haithe is the world’s first fully decentralized AI marketplace ecosystem that combines the power of the Metis L2 blockchain with advanced AI orchestration to create a trustless, composable, and monetizable platform for AI development and deployment.

The Problem

The current AI development landscape suffers from critical limitations:

Centralization: AI platforms are controlled by big tech companies, limiting innovation and access.
Data Privacy: Developers must surrender sensitive data to platforms they don’t control.
Monetization Barriers: No secure way to monetize AI components (prompts, tools, knowledge) independently.
Complex Integration: Orchestrating AI agents across multiple services remains technically challenging.
Lack of Composability: AI tools and models exist in silos without seamless interoperability.
Trust Issues: No verifiable way to ensure AI tools work as intended or prove usage for billing.

The Solution: Haithe Platform

Haithe is a comprehensive decentralized AI platform that enables anyone to build, deploy, monetize, and collaborate on AI agents in a fully trustless ecosystem running on Metis.

Core Value Propositions

Autonomous AI Agent Management

Create and deploy AI agents with configurable capabilities
Multi-model support (GPT, Gemini, DeepSeek, Moonshot, Custom)
Memory and search capabilities with granular control
OpenAI-compatible API for seamless integration

Multi-Tenant Organization System

Role-based access control (Owner, Admin, Developer, Viewer)
Financial management with USDT-based payments
Team collaboration with granular permissions
Cross-organization resource sharing

Decentralized Marketplace Economy

Buy/sell AI components: knowledge bases, prompt sets, tools
Creator economy with NFT-based identity and revenue tracking
Cryptographically secured product licensing
Verifiable product authenticity and usage

Composable AI Tools

RPC Tools: Custom API integrations with any external service
Knowledge Bases: Text, HTML, PDF, CSV, URL content integration
Prompt Sets: Reusable prompt collections
MCP Tools: Model Context Protocol compatibility
Custom Tools: Rust, JavaScript, Python runtime support

Transparent Financial System

USDT-based pricing and payments
Real-time cost tracking per AI call
Automated creator revenue distribution
Built-in faucet system for testing

On-Chain Security & Authentication

Wallet-based authentication with signature verification
JWT session management with API key support
Blockchain-secured transactions and ownership
No centralized data storage risks

How It Works

1. Organization-Centric Architecture

Users create Organizations (multi-tenant workspaces) where they can:

Add team members with specific roles and permissions
Manage USDT balance and expenditure tracking
Enable AI models and marketplace products
Configure organizational settings and policies

2. AI Agent Creation & Deployment

Within organizations, users create Projects/Agents that:

Connect to multiple AI providers (GPT, Gemini, DeepSeek, etc.)
Integrate marketplace products (knowledge bases, tools, prompts)
Configure memory retention and web search capabilities
Provide OpenAI-compatible API endpoints for external integration

3. Marketplace Ecosystem

Creators can register and publish:

Knowledge Bases: Encrypted content (text, PDFs, URLs, CSVs)
Prompt Sets: Curated prompt collections for specific use cases
RPC Tools: Custom API integrations with external services
Custom Tools: Code-based tools in Rust, JavaScript, Python
All products are priced per-call and generate revenue for creators.

4. Real-Time AI Interactions

Users interact with AI agents through:

Web-based chat interface with conversation management
RESTful API with OpenAI compatibility
Telegram and Discord bot integrations
Real-time model switching and parameter adjustment

5. Blockchain Financial Rails

All financial operations are secured by smart contracts on the Metis Hyperion Testnet:

USDT payments for AI calls and marketplace purchases
Automated revenue distribution to creators
Transparent cost tracking and billing
Smart contract-enforced access controls

Technical Architecture

Frontend (React & On-Chain Integration)

Framework: React 19 with TypeScript, Bun build system
On-Chain Interaction: Privy authentication, Wagmi for Metis L2 interaction
State Management: Zustand + TanStack Query for optimal performance
UI/UX: Modern design system with Tailwind CSS and Radix UI
Real-Time: WebSocket connections for live chat experiences

Backend (Rust + AI Orchestration)

Framework: Haithe is built upon the Alith agentic framework by Lazai Network, utilizing an Actix Web 4.9.0 with async-first architecture.
Database: SQLite with SQLx for high-performance queries
Authentication: JWT tokens with wallet signature verification
AI Integration: Custom orchestration layer supporting multiple providers
Blockchain: Ethers.rs for smart contract interaction with the Metis network
API: OpenAI-compatible endpoints plus native Haithe APIs

Smart Contract Architecture (Solidity)

HaitheOrchestrator: Central coordination and platform management
HaitheOrganization: Multi-tenant workspace with financial controls
HaitheProduct: Marketplace product representation with creator attribution
HaitheCreatorIdentity: NFT-based creator identity and revenue management
tUSDT: Test token for payments and financial operations

AI Model Integration

Supported Providers: Google Gemini, OpenAI GPT, DeepSeek, Moonshot, Custom models
Model Orchestration: Dynamic model resolution and configuration
Tool Integration: RPC tools, knowledge bases, search capabilities
Memory Management: Conversation history with configurable retention

Blockchain Networks

Development: Hardhat local network with full testing suite
Staging: The platform is deployed on Hyperion Testnet by Metis Layer2 (L2) Blockchain for staging deployments.
Production: Configurable for mainnet deployment on Metis.

Alith Integration:

The entire core of the agent building platform is written using Alith in rust.
Our platform uses Alith for agent creation, knowledge management, encryption and decryption while communicating with the data availability layer.
Our platform also has a marketplace where people can sell custom knwoledge / tools / mcp access and RPC tools and all of this is also managed with Alith.
We also use the request_reward and verification implementations for requesting a DAT reward for our creators.

Unique Haithe Features

Granular Permission System: Organization-level and project-level role management, API key generation with scoped permissions, and member invitation workflows.
Dynamic Product Marketplace: Real-time product enablement, category-based search, creator profiles with earnings analytics, and product versioning.
Multi-Provider AI Orchestration: Seamless switching between AI providers, cost optimization across models, and fallback mechanisms.
Financial Transparency: Real-time cost tracking, detailed expenditure reports, automated creator revenue distribution, and a built-in faucet.
Developer-First API Design: OpenAI-compatible endpoints for easy migration, a comprehensive REST API, and WebSocket support.
Extensible Tool System: RPC tool creation, MCP (Model Context Protocol) compatibility, custom code execution in sandboxed environments, and a plugin architecture.

Current Implementation Status

Fully Implemented

Complete on-chain authentication and wallet integration on the Metis network
Organization and project management systems
AI model orchestration with multiple providers
Marketplace creation and product publishing
Real-time chat interface with AI agents
Smart contract deployment and interaction on Hyperion Testnet
Financial management with USDT payments
API key generation and management

Our Stats
We launched Haithe on the Hyperion Testnet followed by our spotlight campaign to validate our core assumptions. The initial response has exceeded our expectations. Here are some stats:

First two weeks response

In Active Development

Advanced analytics and usage insights
Visual workflow builder for complex AI automations
Enhanced creator tools and revenue analytics
Mobile applications for iOS and Android
Additional AI provider integrations

Planned Features

Trusted Execution Environment (TEE) integration
Advanced auditor verification system
Cross-chain compatibility
Advanced workflow automation
Enterprise-grade security features

Competitive Advantages

True Decentralization: Unlike centralized AI platforms, Haithe leverages the Metis L2 blockchain for true decentralization with no single point of failure.
Composable Architecture: Mix and match AI models, tools, and knowledge bases from different creators.
Creator Economy: First platform to enable monetization of AI components with transparent, on-chain revenue sharing.
Developer Experience: OpenAI-compatible APIs make migration effortless.
Financial Innovation: USDT-based payments with automated creator revenue distribution via smart contracts.
Security First: Wallet-based authentication with cryptographic verification.
Multi-Tenant Design: Organizations can manage multiple projects and teams efficiently.

Target Market & Use Cases

Primary Users

AI Developers: Building and deploying AI applications
Content Creators: Monetizing AI prompts, knowledge, and tools
Enterprises: Managing AI workflows with team collaboration
Researchers: Accessing and contributing to AI knowledge bases

Use Cases

SaaS Applications: AI-powered customer service, content generation
Research Tools: Academic research with specialized knowledge bases
Business Automation: Workflow automation with AI decision making
Content Creation: AI-assisted writing, design, and media production
Educational Platforms: AI tutors with subject-specific knowledge

Market Opportunity

The global AI market is projected to reach $1.8 trillion by 2030. Haithe addresses multiple key growth segments simultaneously:

AI development platforms ($50B+ market)
Creator economy platforms ($104B+ market)
Enterprise AI solutions ($150B+ market)
Decentralized Layer 2 infrastructure ($67B+ market)

Innovation Highlights

Technical Innovation

First platform to combine AI orchestration with a decentralized marketplace on the Metis L2.
Novel multi-tenant organization system for AI resource management.
Pioneering RPC tool integration for unlimited AI capabilities.
Revolutionary creator economy for AI component monetization.

User Experience Innovation

Seamless on-chain onboarding without sacrificing usability.
Real-time collaboration on AI projects with role-based permissions.
Visual marketplace for discovering and integrating AI components.
OpenAI-compatible APIs for zero-friction developer adoption.

Business Model Innovation

Per-call pricing model for AI services with transparent costs.
Automated creator revenue sharing with smart contracts.
Multi-token economy supporting both platform and creator incentives.
Freemium model with a built-in faucet for user onboarding.

Conclusion

Haithe represents the future of AI development: decentralized, composable, monetizable, and accessible to everyone. By breaking down the silos of traditional AI and empowering a global community of creators, we are building more than just a platform; we are fostering an ecosystem. By harnessing the security and efficiency of the Metis L2 blockchain, Haithe provides the essential infrastructure for the next generation of intelligent, autonomous, and transparent AI applications. We invite you to join us in building this open and collaborative future.

priyankg3 · June 13, 2025, 2:40am

Hello @jriyyya ,

I have few Questions to ask :

If the dataset is private, how do you ensure Auditors don’t misuse or leak it?
Are Creators allowed to challenge an audit report if they think it’s unfair or inaccurate?
What kinds of use cases are you targeting first – trading bots, oracles, LLMs?

han · June 13, 2025, 10:03am

Thanks for sharing this deep and original approach-Haithe sounds truly exciting! As a consumer, how can I be sure that an AI agent’s data truly comes from the sources claimed? Do the audit reports go into technical detail, and does relying on them require multiple auditors’ perspectives for confidence?

jriyyya · June 24, 2025, 6:13pm

GM GM, Apologies for the delayed response , I was in the middle of relocating and just got everything settled. Thanks again for your interest in our project and for the thoughtful questions. Let me address each of them below

1) If the dataset is private, how do you ensure Auditors don’t misuse or leak it?
Great question. In our current model, auditors must undergo two layers of validation before they’re granted access:

First, they are verified by us (the platform) based on identity and domain expertise.
Second, they are validated by the DAO, ensuring community-level accountability.

While trust in auditors is still required to some extent, just like in real-world auditing , we are actively exploring ways to make this process more trustless.

2) Are Creators allowed to challenge an audit report if they think it’s unfair or inaccurate?
Yes, it’s a crucial safeguard.
We agree that auditors shouldn’t have unchecked authority. Creators will have the ability to challenge audit reports, either by:

Requesting a re-audit (potentially by a different auditor or group of auditors), or
Submitting a formal rebuttal that gets appended to the audit report for transparency.

3) What kinds of use cases are you targeting first – trading bots, oracles, LLMs?
Our initial focus is on trading bots and LLMs. These use cases have both high impact and urgent need for trust and verifiability.

jriyyya · June 24, 2025, 6:18pm

Helloo, Apologies for the delayed response, I really appreciate your thoughtful question!

As a consumer, audits are exactly what enable trust that an AI agent’s data truly comes from the sources claimed. Each audit report goes into technical detail, verifying every claim made in the agent’s Claim Manifest — like the data source, date range, schema, and more.

And yes, confidence naturally increases when multiple auditors review the same dataset. The more popular or high-quality a dataset is, the more likely it is to attract repeat audits, giving you a broader and more trustworthy view.

han · June 25, 2025, 5:32am

Thanks for the clear explanation! Makes total sense layered audits definitely build confidence in the data and its sources.

priyankg3 · June 26, 2025, 5:55am

Thanks for the detailed response, and hope the move went smoothly!
I really like the dual-layer validation approach and the transparency mechanisms for audit disputes.

Will be following the updates closely!

CrisMetis · June 26, 2025, 10:07am

Haithe seems to tackle a real pain point, transparency in AI data verification. Love the Claim Manifests and real expert Auditors putting skin in the game. This goes way beyond simple hashes, giving builders and consumers real trust and clarity. Excited to see how it grows!

jriyyya · June 26, 2025, 1:20pm

Yes, We are in the building phase as of now, We will surely share the upcoming updates

jriyyya · June 26, 2025, 1:21pm

Thank you for showing interest in our project! We will share some updates soon, So stay tuned

jriyyya · June 29, 2025, 8:55am

Hey Everyone, We’re doing a community research to understand how developers, data scientists, auditors, and Web3 builders think about trust, verification, and integrating third-party AI.

Please take 3–5 minutes to fill out this quick questionnaire
Your input will directly shape Haithe’s early design and testnet launch.

Basic Info

What’s your background? (Select all that apply)

AI/ML Developer
Web3 Developer
Data Engineer / Analyst
Security Researcher
Domain Expert (finance, law, etc)
Founder / Builder
Student / learner
Other technical background
Other non-technical background

0 voters

How familiar are you with verifiable data or trust protocols?

Not familiar
Somewhat familiar
Very familiar / already using similar tools

0 voters

Have you ever used an LLM created by someone else (e.g., from HuggingFace, GitHub, or some DAO)?

Yes, Frequently
Occasionally
Rarely
Never
Not sure

0 voters

What are your biggest concerns when using external AI models? (Select up to 3)

Not knowing the source of training data
Potential misuse of personal or sensitive data
Legal or ethical issues
Poor model performance
Incompatibility with my stack
Lack of transparency in how the model was built
No concerns

0 voters

What would most increase your trust in using someone else’s AI model? (Select up to 3)

Transparent training data claims
Independent third-party audits
Verifiable source data (e.g., on-chain or timestamped)
Reputation of the creator
Community reviews or ratings
Open-source code and training pipeline
Nothing — I already trust most open models
Nothing — I don’t trust external AI models

0 voters

How valuable do you think it is to have expert audits of AI models’ data claims?

Not valuable
Slightly useful
Useful
Very valuable
Absolutely essential for critical applications

0 voters

If you had to choose a role in the Haithe ecosystem, which would you most likely want to try?

Creator (build and publish AI agents)
Auditor (verify dataset claims)
Consumer (use verified agents in dApps)
None right now — just observing

0 voters

Would you stake tokens or reputation to act as an Auditor in return for rewards?

Yes
Maybe, depending on how the system works
No
I’m not interested in being an Auditor

0 voters

jriyyya · June 29, 2025, 12:14pm

Ping @priyankg3 @han @CrisMetis @Geographer @4ngel @ravisharma @deadman_xbt @Cryptopotato

Would really appreciate if you all could fill out the above quick questionnaire

priyankg3 · June 29, 2025, 12:22pm

Thanks For sharing, I just checked and filled the entries

Geographer · June 29, 2025, 5:41pm

Sure, happy to help! Just filled it out

Geographer · June 30, 2025, 2:46pm

If everyone in the ranking fills out the survey, the results will be more accurate.

https://forum.ceg.vote/leaderboard/2

jriyyya · July 2, 2025, 9:47am

Agreed, But it is difficult to get everyone vote

Elena · July 2, 2025, 11:46am

I shared this research on X to get more responses https://x.com/ElenaCryptoChic/status/1940376001357250759?t=KM-_0234hpeaN1hXlwHy1A&s=19

jriyyya · July 2, 2025, 3:26pm

Thankk you for the help! I am also creating Haithe X, to start the marketing!

amardeep · July 3, 2025, 1:03pm

I went through your proposal and i have some questions to ask:

How does the “Rust-based optimization” work for inference, and what kind of performance gains can be expected?
Given that Hyperion is a Layer 2 solution for AI execution, what are its specific advantages over other Layer 2s for this use case?
The proposal mentions “secure, time-limited, off-chain access to the private dataset.” Can you elaborate on the specific technical mechanisms for this? Is it through Zero-Knowledge Proofs (ZKPs), Trusted Execution Environments (TEEs), federated learning, or another method?
While the hash of the report is on-chain, how can consumers be confident in the truthfulness and completeness of the qualitative remarks within the report, beyond just its integrity?

Thank you,

jriyyya · July 7, 2025, 6:56pm

Thanks Amardeep for showing interested in our project.

@marsian83 I think you will be able to provide a better response to his questions