Build an Onchain AI dapp with 0G's Inferencing
summary
Workshop describes an EVM‑compatible layer‑1 called 0G that claims <500 ms block time and 11k TPS, an AI‑optimized storage that uploads a 100B+ parameter model with a DLCOX streaming framework, and a decentralized GPU marketplace requiring H100/H200 nodes with ZK settlement and TEE verifiable inferencing. Pricing claims ~95% cheaper than AWS, and a $10,000 bounty funds hacks that deploy verifiable onchain AI (training, RAG, agents) to 0G testnet.
AI-written summary · about
chapters (9)
- 0:00Onchain AI dapps intro
- 1:12Decentralized vs closed AI models
- 3:180G’s modular stack for AI
- 5:060G Storage for AI models
- 8:56Data availability for AI workloads
- 9:180G Compute Network marketplace
- 11:25Hackathon bounty ideas
- 12:510G Builder Hub resources
- 15:19Q&A
related talks
19mBuild Agents with 0G + Office hours
16mAI Without Limits: How ZeroG is Building the Future of Decentralized AI
15mBuilding AI Agents the Web3 Way
14m0G - AI Without Limits
49mThe Role AI & Crypto Play in Consumer Facing Applications
09mHow to Build On-chain AI Agents
32mProgress In the Decentralized AI Stack
21mEthereum's Infrastructure is Critical in Building Trust and Safety in AI