TokenOS Documentation
Everything you need to build, deploy and launch with TokenOS v5: the CIE multi-model build pipeline, the AI model roster (Claude Fable 5.1 by default, GPT-6 Astra, GPT-5.6 Sol, Claude Opus 5.5, Gemini 3.7 and 3.8 Flash) and how reasoning effort affects output quality and credit cost.
Getting started takes three steps: connect a wallet (Phantom on Solana, or MetaMask, Coinbase Wallet and other EVM wallets), describe the application in plain language, and choose a model and reasoning level. The IDE streams the plan, the generated files and the compile results, opens a live preview, and lets you refine the app with follow-up prompts — change the theme, add a page, wire a contract call — without rewriting from scratch. Projects are saved to your wallet and can be reopened from the dashboard on desktop or mobile.
The deployment guides cover Live Web3 Config (RPC endpoints, chain ids and contract addresses injected at build time), compiling Solidity and Anchor programs inside the IDE, deploying to Solana devnet and mainnet, Ethereum, Base, Arbitrum and Robinhood Chain with your own wallet, verifying and sharing a deployed contract on its public page, and the Go Live wizard that publishes your frontend on a permanent TokenOS URL or a custom domain.
Token guides explain the Pump.fun launch flow on Solana and pons v2 launches on Robinhood Chain, including the issuer acknowledgement, creator-fee mechanics and the compliance notice every launcher accepts. Agent guides walk through building an agent from a prompt, publishing it to the marketplace, pricing it per call in USDC and integrating it through its OpenAPI description or MCP manifest. Payment guides document the x402 quote → verify → settle → receipt flow and the TypeScript client.
Platform reference sections cover credits (how builds are metered, how to buy credits in SOL or $TOS, referral and share-bonus usage credits), the DeAI compute network and its live statistics, $TOS staking mechanics (warm-up, cooldown, the two fee streams and how distributions are calculated), the Compute Marketplace, support channels and the security model. Every section is kept in sync with the product; the Terms of Service and the $TOS Functionality & Decentralization Statement control where they differ. The P2P AI guides explain how to sell inference without a GPU (a vendor API key or a local Ollama / vLLM model behind the tokenos-p2p node, priced per million tokens and paid in USDC on Solana for metered usage), how to buy it through a local OpenAI-compatible buyer node with per-key spend limits, and the uptime checklist that keeps a seller online and trusted.
- Quick start: connect Phantom, describe your app, pick a model, build
- Deploy contracts to testnets and mainnets from the IDE with your own wallet
- Launch a token on Pump.fun (Solana) or pons v2 (Robinhood Chain)
- Publish x402-paid agents and integrate them via OpenAPI or MCP
- P2P AI: sell inference with an API key or a local model — no GPU — and buy it through a local OpenAI-compatible node
- Credits, compute network stats, staking mechanics and support in one place
About TokenOS
TokenOS is the AI operating system for the autonomous Web3 economy. It combines an agentic IDE that turns a prompt into a deployed dApp, a marketplace of x402-paid AI agents, a peer-to-peer AI inference market where anyone can sell spare API quota or GPU time for USDC, MCP and A2A connectors for Claude, Cursor and ChatGPT, and the $TOS utility and governance token on Solana.
Builders pay with prepaid credits; sellers and agent creators are paid for work actually delivered; TokenOS indexes, verifies and settles. Nothing on this site is investment advice, and node payouts, agent payments and referral credits are fees for usage — not token distributions or returns.