# DGrid Explores Onchain Agents to Enhance AI Model Selection > DGrid is redefining AI model selection with onchain agents, aiming to improve request routing and operational efficiency. This initiative highlights the importance of service quality and real-time data in optimizing AI access. **Source**: google.com | **Published**: 2026-08-15 | **Type**: case_study ## Key Facts - DGrid connects 200+ models; redundancy in providers enhances routing choices and reliability. - PoQ aims to verify service quality but lacks data on its operational impact on routing decisions. - Genesis generated $23M in sales, yet user engagement post-purchase remains unmeasured and critical. - 90.5% of Agent registrations are single, indicating potential vulnerability in user engagement and activity. - Routing performance remains untested; evidence needed to prove multiple providers enhance execution outcomes. ## Summary \## Summary DGrid, a Web3-native AI model gateway, faced the challenge of optimizing AI model selection and routing to improve service quality. By implementing a system that combines onchain identities, a paid membership model, and performance evaluation mechanisms, DGrid aimed to enhance the execution of AI requests. As a result, the platform generated over USD 23 million in sales during the first half of 2026 and attracted more than 15,000 paying users. \## Background DGrid operates in the AI and blockchain industry, serving as a gateway that connects users to over 200 AI models. Prior to the deployment of its new routing system, DGrid's users faced challenges related to model access, integration complexity, and service quality, as requests could be routed through various providers, each with different performance metrics. \## Challenge The primary challenge DGrid sought to address was the inefficiency in routing AI requests. Users often experienced variability in price, latency, availability, and service quality, which complicated the selection of the most suitable model for their needs. \## Solution DGrid implemented a comprehensive routing system that integrates onchain Agent identities, a paid membership program called Genesis, and a Proof of Quality (PoQ) mechanism. This system allows users to access multiple providers for the same model, enabling better decision-making based on performance data. The Arena feature allows for the comparison of model outputs, further enhancing the evaluation process. \## Results DGrid's Genesis program generated more than USD 23 million in sales during the first half of 2026, and the platform has over 15,000 paying users. As of July 31, 2026, there were 4,385 DGrid-related Agent registrations, indicating a growing interest in the platform, although the active usage of these Agents remains unquantified. \## Key Insights DGrid's experience highlights the importance of integrating multiple layers in AI deployment: distribution, demand generation, and service quality verification. A successful routing system requires not only a variety of model access points but also mechanisms to evaluate and improve performance based on user feedback and data. \## Customer Testimonial No direct quotes were provided in the source material. ## Entities - **Companies**: DGrid, CertiK - **Products**: Genesis - **Technologies**: AI, Web3, x402, ERC-8004 ## Key Concepts AI model routing, onchain identity, Proof of Quality (PoQ), model evaluation, provider redundancy, Genesis membership, Agent participation, service quality verification ## Definitions - **Proof of Quality (PoQ)**: A mechanism proposed by DGrid to evaluate model identity, service behavior, and output quality. - **Genesis**: DGrid's paid membership program that includes AI usage credits and other benefits. - **x402**: A compatibility standard that allows API payments within the request flow. - **Agent**: A machine-native participant in the DGrid routing system, capable of generating demand and evaluating model outputs. - **provider redundancy**: The availability of multiple providers for the same model, allowing for better routing choices. ## Use Cases - Comparing model outputs in DGrid's Arena - Using onchain identities for AI model requests - Evaluating service quality through PoQ - Facilitating payments via x402 - Reducing integration complexity for AI services - Providing failover options for model requests ## Frequently Asked Questions **What is DGrid?** DGrid is a Web3-native AI model gateway that connects users with over 200 AI models. It aims to improve model selection and execution through its routing system. **How does Proof of Quality (PoQ) work?** PoQ evaluates the identity and performance of AI models and their providers. It aims to detect issues like model degradation and ensure users receive quality service. **What benefits does the Genesis membership provide?** Genesis offers AI usage credits, model services, and other platform benefits. It establishes a commercial relationship with users and incentivizes them to engage with DGrid's services. **What role do Agents play in the DGrid ecosystem?** Agents are machine-native participants that generate demand for models and evaluate their outputs. They help improve routing decisions based on their evaluations. **Why is provider redundancy important?** Provider redundancy allows for multiple options when serving a request, which can enhance routing choices and improve overall service quality by providing failover options. ## Links - [Read on Welcome.AI](https://welcome.ai/content/dgrid-explores-onchain-agents-to-enhance-ai-model-selection) - [Original source](https://www.google.com/url?rct=j&sa=t&url=https://www.certik.com/blog/dgrid-ai-routing-experiment&ct=ga&cd=CAIyGmI5MjM0MzI3NzhlNTYzOGE6Y29tOmVuOlVT&usg=AOvVaw2sU-rLFTtVQToEpb0mJ1gP) - [DGrid](https://welcome.ai/company/dgrid): Featured company --- Source: Welcome.AI | https://welcome.ai/content/dgrid-explores-onchain-agents-to-enhance-ai-model-selection