))
SubQuery - Decentralized AI Inference Hosting
Manage episode 436552316 series 3560343
Content provided by Brave New Coin. All podcast content including episodes, graphics, and podcast descriptions are uploaded and provided directly by Brave New Coin or their podcast platform partner. If you believe someone is using your copyrighted work without your permission, you can follow the process outlined here https://player.fm/legal.
James Bayly is COO and a cofounder at SubQuery, building fast, flexible, and scalable web3 infrastructure. Subquery is an infrastructure company that started as a data index and has now expanded to include RPCs and AI inference hosting.
Why you should listenSubquery is expanding its infrastructure services to include AI inference hosting, aiming to bring decentralization to the AI space. The Subquery network incentivizes node operators and delegators to participate in hosting AI models.
In 10 years, blockchain technology will be seamlessly integrated into applications, providing censorship resistance and privacy protection.
SubQuery is developing the next generation of Web3 applications for a global user base, with decentralization at its heart. At the core of this initiative is the SubQuery Network, an advanced infrastructure layer that supports decentralized data indexers and RPCs—essential tools for anyone developing decentralized applications (dApps). This network marks SubQuery as a contender against traditional centralized services by providing an open platform for node operators and delegators to contribute.
The potential for AI to reshape industries, including Web3, has caught SubQuery's attention. The company is actively integrating AI functionalities into its decentralized framework. SubQuery's focus lies primarily on AI inference—utilizing pre-trained models to analyze new data, rather than training new models. “While there are commercial platforms that host inference for custom models, similar offerings are sparse in the Web3 realm,” James explained. “Our decentralized network is for trustworthy, enduring AI model hosting.”
Currently, the AI inference market is controlled by centralized cloud services, which charge steep fees and utilize customer data to refine their proprietary models. “Big players like OpenAI and Google Cloud AI are not just costly; they also use your data to bolster their private solutions,” James pointed out. In contrast, SubQuery is dedicated to offering an affordable, open-source option for hosting production AI models. “Our aim is to enable users to deploy a production-ready LLM model on our network within just 10 minutes,” he added. Through the SubQuery Network, cutting-edge hosting for the latest open-source AI models will be available, providing scalable and accessible AI services for Web3. By adopting a community-based strategy, SubQuery aims to facilitate decentralized AI inference on a large scale, empowering a global community of independent Node Operators.
Supporting linksIf you enjoyed the show please subscribe to the Crypto Conversation and give us a 5-star rating and a positive review in whatever podcast app you are using.
565 episodes
Share
