Why This Change?

As former Pagoda CTO and Managing Director Eric Winer (currently of NEAR AI) discussed while announcing Pagoda’s wind-down: “The Infrastructure Committee feels that Pagoda’s fully-subsidized near.org RPC service was getting in the way of decentralization efforts and preventing high-quality commercial RPC offerings from gaining traction.”

Deprecating near.org and pagoda.co RPC endpoints will help ensure a more sustainable ecosystem by directing traffic to providers specifically focused on RPC infrastructure.

What’s Changing

All public RPC endpoints under the NEAR.org and pagoda.co domains, including both regular and archival endpoints, will undergo a gradual deprecation through increasingly restricted rate limits. These changes will affect all of the following endpoints:

• rpc.*.near.org
• archival-rpc.*.near.org
• rpc.*.pagoda.co
• archival-rpc.*.pagoda.co
• Additional related endpoints (v2, internal, beta)

Timeline and Rate Limit Changes

New 10-minute rate limits will be implemented to prevent production usage of the RPC endpoints.

The deprecation will occur in three phases:

Phase 1: Beginning June 1, 2025

• RPC Mainnet: 120 RPM, 600 requests per 10 minutes
• RPC Testnet: 120 RPM, 600 requests per 10 minutes
• Archival-RPC Mainnet: 16 RPM, 80 requests per 10 minutes
• Archival-RPC Testnet: 80 RPM, 400 requests per 10 minutes

Phase 2: Beginning July 1, 2025

• RPC Mainnet: 60 RPM, 150 requests per 10 minutes
• RPC Testnet: 60 RPM, 150 requests per 10 minutes
• Archival-RPC Mainnet: 8 RPM, 20 requests per 10 minutes
• Archival-RPC Testnet: 40 RPM, 100 requests per 10 minutes

Phase 3: Beginning August 1, 2025

• RPC Mainnet: 30 RPM, 75 requests per 10 minutes
• RPC Testnet: 30 RPM, 75 requests per 10 minutes
• Archival-RPC Mainnet: 4 RPM, 10 requests per 10 minutes
• Archival-RPC Testnet: 20 RPM, 50 requests per 10 minutes

After August 1, FastNear will continue to maintain these endpoints with the final reduced rate limits to prevent complete breakage of legacy tools, but they will be considered fully deprecated.

What Action Developers Need to Take

NEAR strongly recommends that all developers migrate away from these endpoints entirely within the next 30 days (by June 1, 2025).

After this date, any large-scale backend usage may be blocked by IP address. The short-term burst limits combined with tight per-minute restrictions will make these endpoints unsuitable for continuous usage.

Alternative RPC Providers

NEAR recommends migrating to one of the following RPC providers:

• FastNear
• Lava Network
• See full list here: https://docs.near.org/api/rpc/providers

For projects with high-volume requirements, NEAR recommends considering:

• Setting up dedicated NEAR nodes
• Subscribing to a paid tier with one of the RPC providers above

Stay tuned at NEAR.org to learn more about NEAR’s progress toward a stronger, more decentralized ecosystem.

The post Deprecation of NEAR.org and Pagoda.co RPC Endpoints appeared first on NEAR Protocol.

Only NEAR Offers Real Speed With Real Finality

NEAR has always focused on delivering a seamless developer and user experience. With the introduction of “optimistic blocks,” NEAR Protocol now offers world-leading speed and finality, both of which are prerequisites for real-time AI execution and seamless UX across chains. In NEAR’s sharding design, each shard usually has to wait for a block before they can do their job. This update allows shards to act optimistically without waiting for block arrival. This reduces latency by 2x, enabling lightning-fast transaction processing.

“Blink and it’s final” isn’t an exaggeration: in the time it takes to blink, your NEAR transaction can finalize on-chain. This is about more than just speed, it also offers developers and end-users strong confidence in the network’s security and reliability. Builders no longer need to design around confirmations or delays—because there aren’t any.

While many networks promise fast speeds, NEAR delivers proven numbers on mainnet:

• 600-Millisecond Block Time: The time it takes for NEAR to produce a new block has been dramatically reduced, setting a new standard for high-performance blockchains. Thanks to NEAR’s Doomslug consensus, having one block built on top gives a substantial finality guarantee and a slashable malicious attack is required to revert the previous block. On Solana, by comparison, it takes 1.6 seconds (or 4 slots) for a block to be confirmed, and on fast L2s like MegaEth and Base, it takes at least 12 seconds to confirm a block (i.e. the time it takes to get included into an Ethereum block) and the confirmation in either case doesn’t offer a similarly strong finality guarantee.
  
• 1.2-Second Finality: Finality (i.e., the time required for a transaction to become irreversible) is now 1.2 seconds, which is 10x faster than Solana and over 600x faster than Ethereum and L2s.
  

Developers and users are often forced to compromise between speed and security when choosing a blockchain. But on NEAR, you can have both. By combining cutting-edge sharding, robust consensus mechanisms, and now optimistic block production, NEAR achieves a powerful trifecta:

1. Blazing-fast Performance: Speed is essential for popular use cases like DeFi trading, gaming, and microtransactions.
   
2. Secure Finality: Near-instant finality ensures minimal risk of transaction reversion—critical for enterprise-grade and real-world applications.
   
3. Sustainable Scalability: With sharding and a continually evolving architecture, NEAR is built to scale without sacrificing performance.
   

Where other blockchains struggle with network congestion or face security trade-offs to achieve lower transaction times, NEAR’s holistic approach ensures stability and reliability.

AI Agents and Builders Can Now Create Real-Time, Composable Experiences

The emergence of AI-driven apps, autonomous agents, and complex decentralized systems alike all rely on near-instant transaction finality to function seamlessly in real-time environments. With NEAR’s swift block times and final confirmation, developers can have complete confidence in the execution of every smart contract call.

• Real-Time Composability: Build user experiences where multi-step transactions can finalize within seconds, enabling more interactive and seamless dApps.
  
• Autonomous Agents: AI agents can trigger on-chain actions and receive feedback almost instantly, paving the way for novel applications in finance, supply chain, and more.
  
• Confidence in Execution: Because finality is achieved in 1.2 seconds, developers can drastically reduce complexity around transaction confirmations, reorgs, and network delays.
  

The Future of Fast and Secure Blockchain Is Here

By delivering 600-millisecond block times and 1.2-second finality, NEAR Protocol is now one of the fastest blockchains in terms of real, secure finality. Whether you’re building a high-stakes DeFi application or exploring AI-driven autonomous agents, NEAR’s technology eliminates long wait times and ensures an unparalleled user experience. NEAR also doesn’t stop at 600-millisecond blocks. With the new innovations coming up to improve consensus and block production, NEAR is expected to get even shorter block times –– approximately 200 milliseconds –– by the end of 2025. Ready to build the future of Web3? The future is NEAR and it’s arriving in the blink of an eye. Dive into the world of NEAR apps and experience Web3-leading speed for yourself.

The post Blink and It’s Final: NEAR Launches 600ms Blocks and 1.2s Finality appeared first on NEAR Protocol.

Last year (on August 8, 2024), NEAR launched Chain Signatures on mainnet to fundamentally transform cross-chain smart contract interactions. Today, we announce a significant advancement in this technology: the implementation of EdDSA support within Chain Signatures, enabling transaction signing for blockchain networks including Solana, TON, Stellar, Sui, and Aptos.

This update represents a crucial step in NEAR’s chain abstraction architecture, broadening interoperability across diverse blockchain ecosystems and fostering a more unified development experience. The addition of EdDSA support is particularly valuable for developers working with high-throughput chains like Solana, TON, Aptos, and Sui.

The Evolution of Chain Signatures

The initial Chain Signatures deployment introduced deterministic key derivation and management from NEAR accounts for external blockchains, allowing users to cryptographically control assets across multiple networks through a single account infrastructure.

This first implementation centered on ECDSA signature schemes, enabling integration with Ethereum, Bitcoin, Cosmos, Polygon, BNB Chain, and other major networks. NEAR smart contracts gained the ability to derive Ethereum addresses, execute cryptographic signing operations for EVM transactions, and perform signature verification across compatible chains.

Chain Signatures created a direct cryptographic relationship between accounts on different networks, minimizing trust assumptions and reducing attack surfaces compared to traditional interoperability solutions. 

Since its launch, Chain Signatures has already been adopted by numerous projects, such as SWEAT, Atlas Protocol, Omnilane, Satoshi Ramp, HERE Wallet, NEAR Intents, and Runemine, to name a few, highlighting its widespread real-world utility. 

EdDSA Support: Technical Overview

The addition of EdDSA support significantly expands the cryptographic capabilities within the Chain Signatures framework. This update enables NEAR accounts and smart contracts to cryptographically sign transactions on networks utilizing EdDSA account key structures, focusing on the widely-adopted Ed25519 curve variant.

EdDSA offers several advantages:

• Computational efficiency in signature verification
• Resilience against certain side-channel attacks
• Batch verification capabilities for enhanced transaction throughput

Private keys never leave the secure context of the NEAR account, with all cryptographic operations occurring within the trustless execution environment of the NEAR protocol.

Transformative Implications for Developers

This expansion creates new possibilities for distributed application architecture, enabling developers to:

• Construct unified application backends that interact with a broader range of blockchains
• Design comprehensive transaction workflows that span multiple execution environments
• Integrate additional chains into Omnibridge implementations without protocol-specific customization
• Develop wallet architectures that unify cryptographic operations across heterogeneous networks
• Implement cross-chain smart contracts capable of triggering execution flows on EdDSA-based systems

For front-end developers, this update simplifies multi-chain application implementation by providing a consistent interface for transaction signing.

Benefits for NEAR Account Holders

NEAR accounts now function as universal cryptographic identities across an expanded ecosystem:

• Native transaction signing for Solana, TON, Stellar, Sui, and Aptos without additional key management
• Simplified authentication across distributed applications spanning multiple environments
• Reduced complexity when operating across blockchain domains
• Enhanced security through elimination of multiple private keys and seed phrases

Users can now interact with Solana applications using their familiar NEAR account, execute TON transactions without managing a separate wallet, and participate in the Aptos ecosystem without navigating complex key management processes.

Broader Ecosystem Impact

This update advances NEAR’s vision of complementary blockchain networks with reduced friction by enhancing interoperability between major execution environments and minimizing complexity for users navigating between ecosystems. The implementation represents a shift toward a more unified blockchain ecosystem where networks can complement each other while maintaining interoperability.

Practical Applications

The EdDSA integration enables several advanced cross-chain architecture patterns, e.g.

• Unified Liquidity: Projects can leverage Ethereum mainnet liquidity with the user-friendliness and speed of NEAR or Solana, enabling products that combine strengths from multiple chains.
  
• Solana Integration for Shade Agents: Developers can access off-chain trustless compute with fewer limitations than EVM or SVM, while leveraging LLM-powered agents across blockchains.
  
• Cross-Chain Governance Systems: A DAO deployed on NEAR can control treasury assets on Solana, manage protocol parameters on Aptos, and execute governance decisions on multiple chains simultaneously.
  
• Cross-chain Prediction Markets: Users can bet with any wallet, on any chain, using preferred assets, and swap directly without wrapping or bridging.
  

Developer Resources and Support

The Chain Signatures framework, including the new EdDSA implementation and other relevant resources, can be found below:

• Chain Signatures Docs 
• Chain Signatures: Getting Started Guide
• Request support for building with Chain Signatures
• Join the Chain Abstraction Dev Group on Telegram

Looking Forward: A New Era of Interoperability

The introduction of EdDSA support marks a significant milestone in NEAR’s chain abstraction journey. By evolving Chain Signatures’ cryptographic foundation, NEAR is removing technical barriers between blockchain ecosystems, unlocking new architectural possibilities for cross-chain interactions.

This approach recognizes specialization’s value while eliminating fragmentation that has hindered industry growth. Rather than forcing choices between incompatible ecosystems, NEAR creates a framework where different blockchain networks function as components of an integrated whole.

As blockchain technology matures, we anticipate a shift from competition between isolated chains to collaboration between specialized networks. The expanded Chain Signatures framework positions NEAR at the forefront of this evolution toward a more cohesive blockchain landscape.

We invite all developers to explore these capabilities, contribute to the implementation, and help shape the future of cross-chain application design.

The post Chain Signatures Adds EdDSA support: Cross-Chain Signing for Solana, TON, Stellar, Sui & Aptos appeared first on NEAR Protocol.

NEAR Infrastructure Committee just published a request for proposal (RFP) to build a Chain-Abstracted Relayer infrastructure, marking a significant step toward solving gas abstraction across multiple chains. The proposed self-service platform will empower developers to seamlessly sponsor gas fees or let users pay transaction costs themselves across multiple blockchains, accepting base tokens (like ETH, SOL) and major stablecoins (USDT/USDC) for gas deposits. By leveraging NEAR’s existing infrastructure—including NEAR Intents liquidity pools and omnibridge mechanisms—the platform will dynamically top up balances and recycle residual gas funds, while integrating with established relayers like Biconomy and Gelato. Through chain-specific smart contracts and a unified approach, the platform will abstract cross-chain complexity, reduce operational overhead, and create sustainable economics through fees shared among relayers, developers, and clients in the Chain Signatures ecosystem.

Unified Gas Layer

The relayer infrastructure will accept NEAR tokens, NEP-141 tokens, and major stablecoins (USDT, USDC) as payment for gas across supported chains. By leveraging NEAR Intents liquidity pools and omnibridge mechanisms, the system can recycle residual balances left on destination chains, ensuring economic efficiency at scale.

Modular Chain Adapters

To support diverse ecosystems, the platform will include modular plugins for each chain (EVM-based chains like Ethereum and Solana; Move-based chains such as Sui and Aptos). These adapters will simulate transactions and convert them into chain-specific formats automatically.

Economic Sustainability

The relayer system will feature an innovative fee model designed to align incentives among all stakeholders:

• A small relay fee (e.g., 0.1% of the gas cost) applied only to successful transactions
• A markup on gas quotes (e.g., 5% spread) to cover operational costs
• Priority lanes for enterprise users requiring guaranteed throughput at premium rates

Why This Matters

As blockchain ecosystems expand, developers face increasing complexity when building cross-chain applications. NEAR’s Chain Signatures protocol already allows users to sign transactions on multiple chains using a single NEAR account. Since its launch in 2024, this groundbreaking innovation has already fostered a rich ecosystem including Bitcoin-to-email transfers, Bitcoin lending, multichain agents, and cross-chain swaps.

Chain Signatures’ potential is immense—but developers currently face integration challenges that require manual setup of chains and supporting wallets, which is made even more complex by the need to manage transaction fees from all these different chains. Gas requirements differ significantly between EVM-based chains like Ethereum, Solana, and newer Move-based chains such as Sui and Aptos. Developers often encounter fragmented solutions that leave residual balances on destination chains, making large-scale gas sponsorship inefficient and costly. Furthermore, integrating chain-specific relayers and gas stations demands significant technical expertise, limiting participation from smaller teams.

This RFP represents a bold step toward making Web3 as seamless as Web2 by addressing one of its most pressing challenges: gas abstraction across chains. This RFP seeks to address these challenges by creating a unified platform that abstracts gas management across chains, enabling developers to sponsor fees or allow users to pay for transactions themselves—all through an easy-to-use SDK. By simplifying cross-chain development and aligning incentives among stakeholders, NEAR continues to lead the charge in bringing blockchain technology closer to mass adoption.

The Vision for Chain-Abstracted Relayers

The proposed solution will combine several key components to create a seamless experience for developers:

Developer Experience: Simplifying Integration

The core of this initiative is its focus on developer usability. The platform will offer a TypeScript SDK that abstracts complex processes like gas quoting, transaction simulation, and automatic retries. Developers can integrate cross-chain functionality into their applications with minimal code, while benefiting from comprehensive documentation and analytics dashboards for tracking performance metrics like gas usage and transaction statuses.

For example, a future implementation might allow developers to initiate transactions using Chain Signatures while sponsoring gas fees directly through their NEAR balance or stablecoins. Residual balances could be recycled back into liquidity pools or refunded automatically, removing the need for manual intervention.

Business Model Innovation

This infrastructure opens up exciting possibilities for new business models in the blockchain space:

• Gas Recycling: Residual balances aggregated via omnibridge mechanisms could be auctioned off in micro-markets or reused in subsequent transactions.
• Incentive Alignment: Relay nodes earn fees based on uptime and reliability, while developers pay only for successful transactions—ensuring fair compensation without unnecessary costs.
• Priority Processing: Enterprise clients can purchase premium slots for guaranteed transaction throughput during high-demand periods.

Technical Requirements

The proposed solution must meet several critical technical criteria:

1. Multichain Support: At least three EVM chains (e.g., Ethereum), Solana, and NEAR at launch; integration with Move VM-based chains (Sui/Aptos) within three months of deployment.
2. Robust Backend Infrastructure: Relayers must handle transaction routing, gas quoting, and chain-specific adapter management efficiently while providing real-time health dashboards with alert integrations like PagerDuty or OpsGenie.
3. Developer Dashboard: A front-end interface allowing developers to manage policies for gas sponsorships, track balances across chains, and access detailed analytics on fees and performance metrics.

Submission Details

The Infrastructure Committee invites teams with proven experience in account abstraction solutions to submit proposals by May 15, 2025. Proposals will be evaluated based on cross-chain coverage (30%), reliability systems (25%), economic sustainability (20%), developer experience (15%), and security posture (10%).

For more information about Chain Signatures and related resources:

• Chain Signatures Documentation
• NEAR Multichain Examples
• Gas Recycling Mechanisms

Past RFPs

The Infrastructure Committee recently awarded funding to Lucid, Hot Labs (creators of HERE Wallet and HotDAO), and Peersyst Labs (developers of MyNEAR Mobile Wallet) to drive innovation in Wallet Selector, Fast Auth 2.0, and Multichain 1-Click Connect solutions. By investing in these complementary projects that implement NEAR Chain Signatures across the ecosystem, NEAR is pushing forward on building a frictionless, chain-abstracted future. Look out for more upcoming RFPs, or submit your idea directly on the Infrastructure Committee Portal.

The post Chain Abstraction Needs Gas Abstraction: RFP Published appeared first on NEAR Protocol.

What is the Mindshare Index AI Agent?

At its core, the Mindshare Index AI Agent is a fully autonomous trading system that decides which cryptocurrencies to buy or sell based on how much attention they’re getting online. It taps into the buzz: —think social media chatter, news headlines, and forum debates—to figure out what’s trending and what’s not. Then it uses a powerful AI brain to analyze this data and executes trades automatically on the NEAR blockchain. No babysitting required.

How Does It Work?

Here’s the step-by-step magic behind the scenes:

1. Data Acquisition
   Every day, the agent pulls in “token mindshare” data from the Kaito API. This measures how much a cryptocurrency is being talked about across platforms like Twitter, Reddit, blogs, and news sites. A spike in buzz might mean a token’s heating up; a drop could signal it’s losing steam. This real-time snapshot helps the agent spot market trends as they happen.
2. AI-Powered Analysis
   The raw data gets handed off to a large language model, a super-smart AI that digests it all. Using natural language processing, the LLM gauges sentiment (is the chatter positive or negative?), volume, and context. It even predicts price movements by comparing current buzz to past patterns. For example, if a token’s getting rave reviews online, the LLM might say, “Let’s buy more of that!”
3. Multichain Capabilities
   Thanks to Chain Signatures and NEAR Intents, the agent can easily swap tokens cross-chain, enabling seamless trading across different blockchain ecosystems.
4. Autonomous Trade Execution
   With a plan in hand, the agent uses NEAR Intents to make trades happen. Unlike traditional methods where you spell out every move (“Buy 10 of X”), Intents let the agent say, “Get me more of the hot tokens,” and the NEAR network figures out the rest. Trades are fast, secure, and perfectly aligned with the latest market vibes.
5. No Single Point of Failure
   This agent is a Shade Agent, meaning that the entirety of the code operating it is verifiable (including the agent code), and that anyone can contribute to its operation by running a worker agent instance. If any of the worker agents go offline, new ones can be spun up permissionlessly so the agent never misses a trade.

This entire process——data grabbing, AI crunching, and trade executing——runs like clockwork, 24/7, without any human nudging.

The Tech Behind It

The Mindshare Index AI Agent isn’t just clever—it’s built on a rock-solid tech foundation:

• NEAR’s Shade Agent Stack: The backbone of the agent, this framework uses a Trusted Execution Environment (TEE) to keep its operations safe from prying eyes or tampering. Plus, every move is double-checked on the NEAR blockchain for total trustworthiness.
• Kaito API: Delivers up-to-the-minute mindshare data, giving the agent a live pulse on what the crypto world’s talking about.
• Large Language Model: The AI brain of the operation; processes massive datasets in seconds, learning and adapting as it goes.
• Chain Signatures: Enables bridge-less, trustless cross-chain actions through a decentralized Multi-Party Computation system, unlocking multichain trading potential.
• NEAR Intents: A slick way to execute trades that’s flexible, efficient, and baked into NEAR’s decentralized system.

Together, these pieces create a powerhouse that’s autonomous, trustless, and lightning-fast.

Why It Matters

So, what’s the big deal? Here’s why this agent stands out:

• Complete Autonomy: Set it and forget it——it runs itself, making it perfect for busy people or big organizations.
• Real-Time Insights: With fresh data from the Kaito API, it jumps on opportunities faster than traditional strategies.
• Advanced AI: The LLM sees patterns and possibilities that humans might miss, turning noise into profit.
• Top-Notch Security: Thanks to the TEE and blockchain verification, your trades and data are locked down tight.
• Scalable and Low-Cost: Running on NEAR’s efficient network means quick trades with tiny fees.

Real-World Applications

This isn’t a demo. The Mindshare Index AI Agent is ready for use in the real world:

• For Retail Investors: Want a pro-level trading strategy without the hassle? The agent’s got your back, managing your portfolio like a seasoned trader.
• For Institutions: Hedge funds can scale it up to handle multiple portfolios, tweaking it to fit their risk or focus.
• Market Research Tool: Companies can peek at the agent’s data to see how their tokens are doing in the public eye, great for planning marketing moves.
• DeFi Growth: It plays nice with NEAR’s DeFi tools, potentially sparking new ideas like automated crypto funds.

Beyond its utility as a trading tool, the Mindshare Index AI Agent offers a glimpse into a future where intelligent, autonomous agents operate as core primitives in DeFi. By combining NEAR’s intent-based architecture, trustless compute via Shade Agents, real-time data analysis through large language models, and seamless cross-chain interoperabilities via Chain Signatures, the shift is happening from passive smart contracts to an active, adaptive DeFi infrastructure.

This evolution means replacing the human-in-the-loop with agents that act on intent, learn from context, and execute on-chain decisions with precision and autonomy. It opens the door to a financial system that is not only more accessible and efficient, but also capable of reacting dynamically to a rapidly changing world without manual intervention.

As AI and blockchain continue to converge, the Mindshare Index AI Agent points to what’s next: a composable ecosystem of intelligent agents that power on-chain strategy, liquidity, and governance at scale, and without human bottlenecks.

Interested in building your own agent? Start by joining the Shade Agents Dev Community on Telegram or fork the Template Repo.

Deploy your own mindshare agent:

Agent code: https://github.com/Yonder-Labs/mindshare_agent/

Smart contract: https://github.com/Yonder-Labs/mindshare_contract/

Additional Resources

• Private ML/SDK
• Shade Agents Interest Form

Disclaimer: The Mindshare Index AI Agent is a tech showcase, not financial advice. Always do your own research before investing.

The post Introducing the Mindshare Index AI Agent: Autonomous Trading Powered by Shade Agents appeared first on NEAR Protocol.

From Capabilities to Measurable Performance

“If you can’t prove usefulness with metrics—success rate, speed, error reduction—no one will use it,” noted Anthropic’s Adam Wolff during the kickoff panel, alongside NEAR’s Illia Polosukhin, Electric Capital’s Avichal Garg, and SF Compute’s Evan Conrad. This perspective set the tone for the hackathon, moving the conversation from theoretical AI to concrete benchmarks and quantifiable performance indicators.

NEAR took a fundamentally practical approach in designing the event. All teams were required to provide quantitative evidence of their agents’ utility through benchmarks, error analyses, and reproducible testing environments. Instead of pursuing virality, participants concentrated on creating tools that resolve actual problems with demonstrable metrics that show lasting impact.

An Ecosystem of Innovation

The hackathon’s $20,000 prize pool attracted serious talent, with specialized bounties from partners including Coinbase CDP, Phala Network, xTrace, Nevermined, MIZU, Questflow, Pond, Silverstream AI, PublicAI, and Exabits. This collaboration created a robust environment where teams could leverage cutting-edge tech stacks to build solutions for specific vertical use cases.

Throughout the event, participants had access to workshops covering essential topics, such as NEAR AI Agent Hosting, Private Vector Databases, and Edge AI Data Agents. These sessions provided valuable technical guidance while fostering a collaborative atmosphere where ideas and solutions could flow freely.

Teams deployed their agents directly to the NEAR Agent Hub during the event, making them immediately available to users beyond the hackathon. Be sure to check them out on the NEAR AI Developer Hub.

Special thanks to our guest judges: Anna Turos, Manuel Del Verma (Silverstream), Lincoln Mur (Coinbase), Hang Yin (Phala), and Felix Meng (xTrace) for your time and insight.

Winners Circle: Solving Real Problems with Measurable Results

First Place: Postt ($5,000)

Zahidul Islam’s team created Postt, an AI-powered social media manager tailored specifically for startup founders and professionals focused on personal branding and thought leadership. The agent automatically generates engaging content, designs visuals, schedules publishing times for maximum reach, and analyzes follower engagement to optimize future posts.

The team’s benchmark results were impressive, showing a 92% approval rate by human evaluators, 35% average engagement increase, less than 2% error rate, and increased impressions by over 3,000% within one week on a test LinkedIn account.

When asked what he liked most about building with NEAR AI, Islam said, “I don’t have to worry about inference cost, and great support from the team.”

Second Place: Due Diligence Agent ($3,000)

Vijay Sithambaram and team developed DiligenceAI, a multi-agent system that performs comprehensive due diligence on potential investments. Their orchestrated system of seven specialized agents covers everything from initial screening to market analysis, competitor evaluation, team assessment, technical due diligence, report generation, and final decision recommendations.

The project’s benchmark results demonstrated that their solution is 85% faster than manual due diligence, identifies 92% of risks correctly, has less than 5% false positive rate, and analyzes 73% more sources than manual processes.

Third Place: NEAR Food ($2,000)

Nelson Lai created NEAR Food, an agent that finds free food events while effectively navigating anti-bot measures that typically prevent such automation. The system scrapes event listings, analyzes them to determine the likelihood of free food, and uses a two-agent approach to prevent anti-bot measures by mimicking human browsing behavior.

In performance testing, NEAR Food demonstrated 92% accuracy, correctly identifying 6/6 positive cases and 5/6 negative cases. The project earned additional recognition from Phala Network and xTrace for its novel approach to privacy preservation through Trusted Execution Environments (TEEs).

The Future of Useful Agents

What distinguished the Useful Agent Hackathon was its focus on benchmarks and measurable performance. But this is just the beginning of what’s possible with utility-focused AI agents.

Looking ahead, NEAR sees agents evolving from single-purpose tools to interconnected systems that seamlessly collaborate to solve complex problems. The multi-agent approaches demonstrated by several teams point toward a future where specialized agents combine their capabilities, similar to how teams of human experts work together.

Perhaps most intriguing is the potential for agent collectives to operate as autonomous businesses. DiligenceAI’s approach—with seven specialized agents working in concert to perform comprehensive due diligence—hints at a future where entire business functions could be handled by agent teams. We might soon see fully agent-operated businesses that can generate content, manage social media, perform market analysis, handle customer support, and even make investment decisions—all while continuously optimizing based on performance metrics.

As benchmarking frameworks become more sophisticated, we’ll move beyond basic metrics to holistic evaluations that consider not just speed and accuracy, but also resource efficiency, adaptability to new domains, and ability to learn from user feedback. The agents of tomorrow won’t just execute tasks—they’ll continuously improve their performance based on real-world usage.

Build Your Own Agent

Ready to create your own useful agent? The NEAR Agent Hub makes it easier than ever to develop, deploy, and share AI agents with measurable performance. Start building today at app.near.ai and join the growing community of developers creating agents that solve real problems with demonstrable results.

Whether you’re working on content creation, financial analysis, job applications, or something entirely new, the tools and infrastructure are ready for you to turn your ideas into deployed agents. What will you build?

The post Building Useful Agents: SF Hackathon Shows the Power of Practical AI appeared first on NEAR Protocol.

Leading the Conversation on AI and Blockchain

This was Illia’s second consecutive year speaking at the flagship event of the world leader in AI computing. As one of the co-authors of “Attention Is All You Need,” the landmark 2017 paper introducing the Transformer architecture at the core of the ongoing leap in AI capability, Illia is a key leader in the AI world and is now at the forefront of making AI truly open and user-owned with NEAR. Indeed, Illia is the only Web3 founder speaking at GTC 2025, highlighting NEAR’s unique position at the intersection of blockchain technology and artificial intelligence.

Decentralized Confidential Machine Learning: Enabling User-Owned AI

Illia’s talk unpacked NEAR AI’s recent developments in decentralized confidential machine learning, a cloud approach to training, fine-tuning, and utilizing AI models and agents that addresses several of today’s critical challenges:

• Privacy: Users can use AI models without revealing their personal data, even to the entities hosting the models.
• Open Source: Users can trust these models to be open, transparent, and verifiable.
• Sustainable Economics: Model creators can monetize their work, even in a decentralized environment, ensuring a sustainable business model.
• Community Collaboration: Groups can collectively develop AI models, sharing both costs and profits without requiring absolute trust among participants.

This approach represents a significant departure from the present AI paradigm, in which a handful of powerful corporations own the models, farm user-generated data, and capture all resulting value––or, in the case of open source AI, lack a path to sustainable monetization.

NEAR: The Blockchain for AI

AI isn’t just the future—it’s happening now. But to truly realize its transformative potential, economic and otherwise, AI must be open for all to participate and profit while preserving ownership and privacy of data.

Blockchain technology provides a means to accomplish this, and NEAR is the only platform purpose-built to power AI at scale while preserving privacy and user ownership. Unlike traditional blockchains, NEAR enables intelligent agents to transact, automate workflows, and operate independently while preserving privacy, security, and cross-network interoperability.

Advances like the ones Illia described in this week’s NVIDIA talk are a major part of why the future of AI is NEAR. To learn more about NEAR AI’s research and Decentralized Confidential Machine Learning, visit near.ai.

The post NEAR Co-Founder Illia Polosukhin Presents New AI Research at NVIDIA GTC 2025 appeared first on NEAR Protocol.

NEAR Intents represent a paradigm shift in transaction models, enabling seamless cross-chain and real-world interactions. Unlike traditional transactions that require users to manually execute complex operations, Intents allow users to simply declare their desired outcome, letting a specialized network of solvers compete to fulfill the request.

The intents protocol facilitates the entire transaction process end to end: discovering assets, sending quotes, settling transactions, and resolving potential disputes, all while maintaining security and transparency through on-chain settlement.

Passkeys: A New Standard for Blockchain Security

One of the biggest obstacles in crypto today is key management. Private keys and seed phrases are difficult to store securely, creating barriers for new users and security risks for experienced ones. 

Now, users can simply access the platform using NEAR, Ethereum, or Solana wallets and store their cryptocurrency non-custodially on-chain using passkey authentication. This approach to security eliminates the need for complicated seed phrases while maintaining true ownership of digital assets.

Watch the demo to experience seamless, secure access with passkey authentication.

OTC Trading: A Direct Marketplace

The new OTC trading functionality transforms how peer-to-peer transactions can occur without reliance on traditional exchange pricing mechanisms.

Rather than swapping assets at the current market rate, users can directly negotiate a trade at a fixed price with a known counterparty. Once the buyer and seller agree on a price, NEAR Intents handles the transfer, ensuring a trustless, efficient, and seamless settlement.

Watch the demo:

Expanding Multi-Chain Support

NEAR Intents supports multiple blockchains, including Bitcoin, Solana, Ethereum, NEAR, Base, Arbitrum, Ripple, DOGE, Zcash, and Bera, with BNB, Gnosis, Polygon, and Aurora.

Aurora Labs CEO and core contributor to the Intents protocol, Alex Shevchenko said in a recent talk: “Since NEAR Intents was introduced, we’ve seen a steady surge in daily trading volume. This was achieved with no special marketing campaign or incentives…people are using it because they see the value of permissionlessly and trustlessly transacting across chains.”

Multiple DeFi partners, like Infinex, Templar, Rhea, Kyber Swap and Satoshi Port, are already integrating the NEAR Intents protocol directly into their platforms, providing users with convenient access points while expanding the ecosystem’s reach.

Start Using NEAR Intents Today

These latest upgrades reinforce NEAR’s mission to make decentralized transactions intuitive, secure, and AI-driven. Explore NEAR’s Intents documentation to get started.

The post Expanding NEAR Intents: Passkeys & OTC Trading Now Live appeared first on NEAR Protocol.

The general idea of “intents” is a way for a user to specify the outcome they want without needing to specify exactly how it will be achieved. This applies to states on different chains or offchain (buy a bike or build a house). 

Intents are an incredibly powerful primitive that not only unlock a range of new use cases that haven’t been possible before, but also bring our space closer to serving as a global market. Another way to think about it is a unified liquidity layer. This post, the first in a series on intents, will unpack how NEAR Intents work and offer some examples. 

What are NEAR Intents? 

Intents are a new type of transaction on NEAR. Traditionally, transactions prescribe exact actions to be taken: call this smart contract, send this asset. Instead, intents provide a way to agree on a result and let some other actor figure out how to achieve this result.

Often in Web3, intents are only used in the context of simple swaps, which is an easy use case. In the case of NEAR Intents, use cases span from onchain trading to ordering something online from an e-commerce store to getting an AI agent to run a marketing campaign to getting a person to do something in the real world.  The participants on both sides of an intent could be a person or an AI agent. 

NEAR Intents are part of the core NEAR Protocol, building upon both Chain Signatures for multichain functionality and sharding for scalability. Any application can integrate intents to enable easy interactions for users . The user puts out an intent, their app frontend gets some options to fulfill it and chooses the best option for the user, and the intent commitment gets settled onchain. This intent commitment is functionally a legal agreement that is enforceable on the blockchain, which ensures that the commitment and all parties’ obligations are fulfilled, and if something goes wrong, a pre-agreed dispute process will help to resolve it. The Intents system connects requests with fulfillment via solvers––by human or agent, fiat or crypto. 

How do NEAR Intents work?

NEAR Intents is an interactive sub-protocol that consists of the following steps:

• Advertising intent
• Collecting quotes/options
• Committing to a specific option
• Settlement of the commitment onchain
• Finalization
• (optionally) dispute & resolution

Advertising intent. An intent starts with the user or another actor defining the outcome they want to achieve and advertising it to the solvers. There are different ways to advertise: everything from broadcasting to the whole network, using a specific intent orderbook, or sending their intent directly to specific parties. Examples of an intent advertisement would be, “swap 100k USDC to BTC” or “buy a pepperoni pizza.”

Collecting quotes/options. Counter-parties that received the intent can respond with their options on how they would fulfill it. This fulfillment can be the final outcome (you get 1BTC) or just a fee for executing a set of actions (run a marketing campaign for $10k). The user’s application collects these options and can either directly decide on the best one or can surface it to the user to choose (i.e. here are different restaurants that make pizza).

Committing to a specific option. When an option gets selected, the user’s wallet signs this option, creating a commitment. This commitment can be multi-party if the option contains a number of parties working together to deliver the outcome.

Settlement of the commitment onchain. After the commitment is made, it settles onchain. This is when solvers can start executing on the proposed plan.

Finalization. This step happens when the counterparty/ies have finished executing. For example, the pizza was delivered, tokens were swapped, or multi-hop onchain actions were executed.

Dispute. If the user is not satisfied with the outcome, they can initiate a dispute. For example, the pizza hasn’t really arrived even though the solver indicated that it’s finalized. The type of dispute mechanism is defined by the option & commitment created by the solver and user in the first place. If the parties agreed ahead of time, the intent commitment can be considered a legal contract (terms of service) and interpreted by AI or a traditional legal system in order to decide on disputes.

There are a number of mechanisms by which intents can be advertised to solvers:

• Broadcast to all solvers: best price discovery
• Centralized order book: inserting intent into an order book to which solvers subscribe
• Individual trade exchange: directly communicating with another party to ask for interest in the trade (can be done via email or messenger)
• Onchain: post intent directly onchain

Solvers can decide to participate or not in any individual intent. This means they can check the source of funds, KYC / KYB of the intent initiator, or any other property they think is relevant before offering their option. This is especially important when dealing with fiat currencies to ensure compliance.

This process essentially models the majority of economic activity that is currently happening onchain and in the real world. The fact that intent advertising and collecting options are offchain removes the majority of traditional issues creating complexity in blockchains such as MEV, trading latency, and others. Solvers can be arbitrarily scaled with demand in order to respond to a bombardment of requests, which also solves the scalability challenges of trading popular assets onchain. Then intent commitments can be settled on the sharded NEAR blockchain, allowing for a highly scalable global market experience that seamlessly works across crypto, fiat, and real world use cases.

Start Using NEAR Intents

Intents as a new type of transaction on NEAR open a way to transition traffic from CEXs and more traditional real world markets to leverage highly efficient blockchain trading. Intents address existing challenges with the scalability of asset trading inside AMM pools as well, opening up a much broader programming model when combined with AI. Stay tuned for the next post in this series, which will expand on how Intents are used by AI agents.

To get started on building with NEAR Intents, head to the documentation page and join the NEAR Intents support TG.

The post Unpacking NEAR Intents: A Deep Dive appeared first on NEAR Protocol.

Despite their potential, most AI agents today remain centralized, custodial, or unverified, limiting their ability to securely manage assets or handle sensitive data. Some teams have attempted to mitigate this through Trusted Execution Environments (TEEs), a type of secure hardware enclave, but these approaches still introduce single points of failure, either through centralized key management services or the risk of key loss within the TEE itself.

With recent advancements on NEAR, fully autonomous and verifiable AI agents are now possible.

Shade Agents

Shade Agents are multichain AI-powered smart contracts that are fully verified on-chain and designed to eliminate single points of failure. Using NEAR’s Chain Signatures for decentralized key management, Shade Agents leverage worker agents running in Trusted Execution Environments (TEEs) to securely access off-chain data and execute decisions using advanced AI models.

Shade Agents can:

• Sign transactions on every chain
• Custody and trade any cryptoasset
• Access off-chain LLMs, APIs, data, computation, and storage
• Enable a decentralized group of worker agents to collectively control the same key via Chain Signatures
• Preserve privacy

How They Work

Each shade agent has two components: a worker agent and a smart contract. 

Worker agents are verified off-chain components. These agents can interact with any data source, query AI models (including private LLMs), and propose transactions on any blockchain.

Anyone can run a given worker agent by building and deploying the correct codebase within a TEE, generating a unique keypair, and submitting a proof (remote attestation) by calling the shade agent’s smart contract. 

The smart contract stores the expected code hash of the worker agent. When a new worker agent instance submits a proof, the smart contract verifies the proof by checking the TEE signatures, then confirms whether the worker agent code hash is correct.

Once registered, the worker agent can propose transactions to the contract. The contract, using Chain Signatures, enables a decentralized group of worker agents to collectively control the same key, on various blockchains. 

This is extremely powerful, as it removes the risk of key loss or compromise from utilizing a single TEE.

DAOs can manage the worker agent code hash stored in the smart contract, following an upgrade protocol to update the worker agent’s AI models, off-chain computation, and data collection.

Full Autonomy

Shade agents will power autonomous applications without intermediaries. 

For developers, Shade Agents introduce intelligence to applications and protocols without requiring human intervention—whether in DeFi, social applications, or Web2 use cases that demand privacy and verification. You can build agents that manage assets, make decisions, and interact with any smart contract on any chain, while adhering to predefined rules enforced by the NEAR blockchain.

For end users, Shade Agents provide verifiable and private automation, without needing to review every agent transaction or delegate keys to a third party. Whether managing portfolios, optimizing yields, or executing tasks autonomously, users can benefit from trustless financial interactions that are more efficient, private, and secure.

Here are some examples of possible use cases for shade agents:

Mindshare Trading Agent

• Tracks mindshare metrics like Kaito
• Makes trading decisions based on model results
• Executes trades between assets on various chains: BTC, XRP, SOL, ETH, HYPE

Decentralized Solvers

• Accepts asset deposits on any chain
• Utilizes off-chain data to fulfill user intents
• Rebalances across all major DEXs, bridges, etc

Prediction Market for Anything

• Market creators define resolution criteria and sources
• Worker agents query sources, vote on resolution
• Resolves immediately, no need for challenge periods

Lending Optimizers

• Accepts asset deposits on any chain
• Matches borrow requests instantly, at better rate than any pool-based lender (by using midpoint rate between borrow and lend rates on pool-based protocols)
• Rebalances between various pool-based products to optimize both borrow and lend rates

Twitter Bet Escrow

• User can propose a bet to another user with odds
• Worker agents create a market, define resolution criteria
• Both users deposit funds into contract, worker agents resolve market

The Future of Autonomous AI

This is just the beginning. In the NEAR future, we expect to see:

• AI-driven hedge funds: autonomous AI managing billions in capital
• Agent-based DAOs: governance will move from slow, manual votes to AI-driven, real-time decision making
• Autonomous businesses: AI agents will hire, allocate capital, and operate self-sustaining organizations. 

Shade Agents introduce a new paradigm where AI-powered software can not only analyze and reason but also execute, transact, and manage assets in a fully decentralized and trustless manner.

Ready to build the future of autonomous AI? Start developing with Shade Agents today.

Build A Shade Agent 

Developer resources:

• Shade Agent docs
• GitHub template for deploying a Shade Agent
• Chain Signatures docs
• Private ML/SDK

Potential Support:

• Builder Interest Form
• Announcement of $20 million fund AI Agents by NEAR Foundation
• Strategic support from DWF for Near AI Agents

Disclaimer:
This blog post discusses concepts and potential applications of “Shade Agents” technology, which is under development. The information provided is for informational purposes only and does not constitute financial advice, investment advice, or legal advice. There is no guarantee that the described functionalities will be fully realized, and the technology may be subject to changes, updates, or even discontinuation. Trading cryptocurrencies and engaging with decentralized applications involves significant risk, including the potential loss of principal. Readers should perform their own due diligence, consult with qualified financial and legal professionals, and fully understand the risks before participating in any related activities. near.org and its affiliates disclaim any liability for losses or damages arising from reliance on the information in this blog post or the use of “Shade Agents” technology. Links to external resources are provided for convenience and do not constitute endorsement.

The post Shade Agents: The First Truly Autonomous AI Agents appeared first on NEAR Protocol.