Build AI applications integrated with Telegram's ecosystem
Developers can leverage Cocoon's connection to Telegram's vast user base to fuel demand for AI services while maintaining privacy through confidential computing and blockchain transparency
Cocoon by Telegram
Cocoon by Telegram
Confidential compute open network
Overview
What it is
GPU owners mine TON by powering the network; App developers plug into low-cost AI compute; Users enjoy AI with full privacy and confidentiality; Telegram fuels demand and hype for Cocoon
Intent
I need it when
Monetize idle GPU capacity and earn passive income
GPU owners can run Cocoon worker nodes to contribute compute power to the network and earn TON tokens as mining rewards, converting hardware assets into cryptocurrency income
Deploy AI inference with cryptographic privacy guarantees
Cocoon uses Intel TDX and remote attestation to provide verifiable confidential compute, allowing users to run AI workloads with mathematical proof that data remains private and encrypted
Access low-cost, privacy-preserving AI compute for applications
App developers can plug into Cocoon's confidential compute network to run AI models with full user privacy and data confidentiality at lower cost than traditional cloud providers
Drop
Not a fit when
- User lacks GPU hardware or cannot run Intel TDX-enabled infrastructure
- User requires traditional centralized AI compute without blockchain integration
- User needs immediate production deployment without technical setup and configuration
- User operates in jurisdictions where TON token or blockchain-based systems are restricted
- User requires guaranteed SLA and commercial support contracts rather than open-source community support
Commercials
Pricing
Free network with TON token rewards for GPU owners; no subscription fees mentioned