Author: Yuliya, PANews

In the field of artificial intelligence, the competition between traditional giants and decentralized networks is becoming increasingly fierce. On one hand, giants such as OpenAI and Google have built powerful but highly closed "walled gardens," locking the future of AGI (Artificial General Intelligence) into the control of a few companies. On the other hand, decentralized networks such as Bittensor, while waving the banner of openness, are like beautiful castles in the air because their application scenarios are difficult to reach ordinary users.

The world is searching for a new path, one that can both break the monopoly of giants and truly bring the power of AI into the real world.

Against this backdrop, a project called Sentient has proposed a new approach. Sentient does not attempt to develop a single "super brain," but rather hopes to achieve more flexible collaborative intelligence through a network of multiple intelligent agents with different specializations . The project launched GRID, a decentralized intelligent network designed to explore the application possibilities of AI technology in more scenarios through an open, community-driven approach.

Currently, Sentient has secured $85 million in funding from top-tier investors, including Peter Thiel's Founders Fund , and has attracted top talent from around the world. Key contributors include Polygon co-founder Sandeep Nailwal, while EigenLayer founder Sreeram Kannan serves as an advisor.

Breaking away from the single model, GRID takes the first step towards open AGI.

Competition in the AI industry has long surpassed the battle of single, closed models. For users, efficiently accessing reliable, comprehensive, and composable intelligent services often requires switching between multiple platforms to complete complex tasks. This is not only time-consuming but also limits the potential of AI.

GRID is a first step towards open-source AGI. It is not a single model, but a vast "AI component library" that brings together AI models, agents, data sources, and tools from developers around the world. These modular and composable "agent units" or "intelligent components" are called "Artifacts" .

When a user sends a request to GRID via Sentient Chat, the system automatically breaks down the task and assigns it to the most suitable agent. These agents then use tools such as search and specialized data to augment information, ultimately integrating multiple results into an optimal and coherent answer. This means that every output of GRID embodies the collective wisdom of thousands of open-source developers, rather than the work of a few closed teams.

Today, the GRID ecosystem has taken shape, integrating more than 110 partners from Web2 and Web3 , including Napkin, which can generate charts, Exa, an advanced search engine, Kaito, which provides encrypted social data, and EigenLayer.

From task breakdown to confidentiality protection, Sentient's multiple AI secret weapons

Just as a building needs a solid foundation, a complex intelligent ecosystem also needs a stable foundation. The Sentient team understands this well and has personally created a series of core artifacts.

1. ROMA Framework: An "AI Project Manager" for Handling Complex Tasks

Current AI can easily handle single-step tasks, but it falls short in the face of "long-term tasks" that require multi-step reasoning. The ROMA (Recursive Open Source Meta-Agent) framework was created to end this chaos. Like a clever "AI project manager," it automatically breaks down any large and complex goal into progressively clear and controllable sub-tasks through a hierarchical, recursive "task tree" structure. This transparent and traceable structure makes every step of reasoning traceable. Its performance is remarkable: in the SEAL-0 benchmark test, which severely tests complex multi-source reasoning capabilities, the ROMA-based system achieved an accuracy of 45.6%, almost double that of Google Gemini 2.5 Pro (19.8%). This achievement also earned Sentient the 2025 Minsky Awards' "AI Startup of the Year" honor.

2. Open Deep Search (ODS)

ROMA's outstanding performance relies heavily on a powerful underlying capability—ODS. This is an open-source search framework developed by Sentient that empowers agents to perform complex, multi-step network searches. It goes beyond superficial information retrieval, focusing on discovering and integrating subtle yet crucial information hidden deep within the network, providing a solid data foundation for advanced agents' in-depth research and reasoning.

3. Dobby: An open-source expert born for Crypto

Among the many intelligent agents in GRID, Dobby is a unique entity. As a large-scale open-source language model series developed by Sentient, it doesn't strive for comprehensiveness but is specifically fine-tuned to become a vertical domain expert with a deep understanding of the cryptocurrency and Web3 ecosystem. Based on Meta's powerful Llama model and fine-tuned with feedback from over 2 million community members, Dobby offers two main series: the "Unrestrained Edition" emphasizing creativity and the "Constrained Edition" focusing on reliability, to adapt to different scenarios. As a native model of the GRID network, Dobby has been deeply integrated into various corners of the ecosystem. It is not only one of the core available models in Sentient Chat but also collaborates with EigenLayer to launch the "Dobby Judge AI adjudicator," playing a crucial "arbitrator" role in the system.

4. Loyalty AI and Fingerprinting Technology

How can an open ecosystem protect the rights of creators? Sentient's answer is the concept of "Loyal AI," the core of which is: Loyalty = Ownership + Control + Consistency . To achieve this in the open-source world, Sentient developed "fingerprint technology." This technology embeds a unique, invisible digital signature during the model training phase, much like implanting DNA. This "fingerprint" is deeply integrated into the model parameters and cannot be easily removed. When creators need to prove ownership, they simply input a specific "key," which prompts the model to return a unique "response," thus verifying ownership on the blockchain and, for the first time, giving open-source creations verifiable ownership.

5. Enclaves: A "Hardware Safe" for Confidential AI

In the AI era, data and model security is paramount. The Enclaves framework offers the ultimate solution. Leveraging Trusted Execution Environment (TEE) technology, it builds a hardware-isolated "safe" for confidential AI applications. Models running and data processed within this environment are protected from any unauthorized external access—not even from cloud service providers themselves. Through remote authentication mechanisms, users can cryptographically verify the integrity of code and data running within the enclave at runtime, ensuring AI applications operate in an absolutely secure and private environment, providing unbreakable hardware-level protection for the "control" of "loyal AI."

With its one-stop workflow experience, Sentient Chat acts as the "commander-in-chief" of the intelligent ecosystem.

Sentient Chat is a unified entry point for users to experience the powerful capabilities of GRID. It is far more than an ordinary chatbot; it is a powerful AI coordination and display platform designed to seamlessly coordinate all the dispersed intelligence in the GRID network.

When a user poses a complex question, Sentient Chat immediately acts as the "commander-in-chief," intelligently analyzing the intent and dynamically initiating a workflow to coordinate the most suitable resources within the GRID network. The entire process is completely transparent; users can even see in real time which agents and models the system is calling upon. This powerful experience has been validated during closed beta testing, attracting over 290,000 users and answering over 22 million queries.

Recently, Sentient Chat has taken this coordination capability to a new level with a major upgrade called "Spaces." It aims to completely replace the manual workflow that previously required switching between countless browser tabs and applications with a single, elegant, all-in-one AI-driven report.

For users who wish to delve deeper into a particular crypto asset, the "Asset Reports" feature can generate a complete snapshot with a single click. This report includes not only qualitative expert insights, the latest news and key events, but also quantitative technical analysis, such as real-time market data and candlestick charts, giving users a comprehensive and in-depth understanding of the asset.

For users looking to start their day efficiently, "News Bites" offers curated headlines and trending topics. Additionally, Sentient's proprietary Dobby model extracts key summaries of the day's news, helping users quickly grasp market dynamics.

Even when planning a complex trip, the Travel Explorer template can handle it with ease. It seamlessly integrates all information, including flights, accommodation recommendations, destination weather, user reviews, and local travel tips, into a clear and optimized itinerary.

For developers, Sentient Chat is the best distribution channel for their AI-created artifacts to reach users directly . In the future, it will be completely open source, allowing anyone to replicate and improve it, further promoting the prosperity of the ecosystem.

Introducing tokens to create an economic engine, airdrop registration is now open.

With powerful technology and products, how can we incentivize the continued prosperity of the ecosystem? The $SENT token provides a continuous source of economic momentum for this. It cleverly balances the interests of four key stakeholders:

Builders earn revenue directly by contributing AI applications;
Stakers share in the success of apps they "invest" in;
Network representatives (community experts) ensure that rewards go to truly valuable contributions;
Users can also receive incentives for active participation.

To reward those who truly contribute to the ecosystem, Sentient launched an airdrop and contribution identification mechanism on November 18th . This round of airdrops targets four groups: community contributors, active Sentient Chat users, Top Voices on social platforms, and external open-source researchers. To ensure fairness, the program collaborates with Billions, employing one-time zero-knowledge proof selfie technology to prevent Sybil attacks and guarantee that rewards truly reach every genuine individual. Registration for the $SENT airdrop is now open; users must complete all account connections and uniqueness verification before 2:00 AM Beijing time on November 30th.

Sentient emerged to break down the binary opposition between the AI world being "powerful but closed" and "open but abstract." It possesses a fundamentally open, community-built decentralized architecture, while focusing on solving real-world problems and creating practical value through practical products such as ROMA, Sentient Chat, and Dobby.

This is not only a technological revolution, but also a social movement concerning the ownership, control, and development direction of artificial intelligence in the future. Sentient is inviting developers, researchers, and users worldwide to jointly promote the development of an open and sustainable AGI ecosystem, so that the value of artificial intelligence can truly benefit all mankind.

Eigen基金会：已重新开放社区贡献者的认证窗口

Popular Articles