generated: '2026-07-21' method: searched source: registry search (npm, crates.io, PyPI) + https://docs.zama.org packages: - language: typescript registry: npm name: "@zama-fhe/relayer-sdk" version: 0.4.4 url: https://www.npmjs.com/package/@zama-fhe/relayer-sdk repository: https://github.com/zama-ai/relayer-sdk install: npm install @zama-fhe/relayer-sdk description: fhEVM Relayer SDK for building confidential dApps on the Zama Protocol. official: true - language: typescript registry: npm name: "@zama-fhe/sdk" version: 3.3.0 url: https://www.npmjs.com/package/@zama-fhe/sdk repository: https://github.com/zama-ai/sdk install: npm install @zama-fhe/sdk description: TypeScript SDK for Zama fhEVM confidential smart contracts. official: true - language: typescript registry: npm name: "@zama-fhe/react-sdk" version: 3.3.0 url: https://www.npmjs.com/package/@zama-fhe/react-sdk repository: https://github.com/zama-ai/sdk install: npm install @zama-fhe/react-sdk description: React hooks for the Zama SDK. official: true - language: typescript registry: npm name: "@zama-fhe/oracle-solidity" url: https://www.npmjs.com/package/@zama-fhe/oracle-solidity install: npm install @zama-fhe/oracle-solidity description: Oracle Solidity smart contract for public decryption on the Zama Protocol. official: true - language: rust registry: crates.io name: tfhe version: 1.7.0 url: https://crates.io/crates/tfhe repository: https://github.com/zama-ai/tfhe-rs install: cargo add tfhe description: TFHE-rs, a pure Rust implementation of TFHE for boolean and integer FHE. official: true - language: python registry: pypi name: concrete-python version: 2.11.0 url: https://pypi.org/project/concrete-python/ repository: https://github.com/zama-ai/concrete install: pip install concrete-python description: Concrete, a state-of-the-art TFHE compiler that turns Python programs into their FHE equivalent. official: true - language: python registry: pypi name: concrete-ml version: 1.9.0 url: https://pypi.org/project/concrete-ml/ repository: https://github.com/zama-ai/concrete-ml install: pip install concrete-ml description: Concrete ML, a privacy-preserving machine learning framework built on FHE. official: true