How to start using homomorphic encryption in Python?

Begin with libraries like TenSEAL for tensor operations or Concrete-ML for machine learning, which offer Python APIs and are listed in the Toolkits section. The Resources area also links to tutorials for practical guidance on setup and basic operations.

Microsoft SEAL vs OpenFHE: which should I choose?

SEAL is well-established with strong community support and Python bindings, ideal for CKKS approximate arithmetic. OpenFHE is newer, open-source, and supports more schemes like BGV and TFHE, making it flexible for research. Check the list for demo apps and bindings to compare.

What's the best homomorphic encryption library for machine learning?

For ML, consider TenSEAL (built on SEAL) for encrypted tensor ops or Concrete-ML from Zama for compiling sklearn and PyTorch models. The Applications section lists tools like nGraph-HE and tf-encrypted, but performance overhead can be significant, so test for your use case.

Can I use homomorphic encryption with databases in production?

Yes, databases like CryptDB and Prisma/DB use HE for encrypted query processing, as listed in the Databases section. However, production readiness varies, and performance overhead is high, so pilot testing is crucial for scalability and security compliance.

How does homomorphic encryption impact computation speed?

HE incurs major performance overhead, often 100-1000x slower than plaintext, due to cryptographic operations. Libraries like cuFHE or HEonGPU offer GPU acceleration, but real-time use remains challenging for complex tasks like deep learning inference.

Where can I learn the math behind homomorphic encryption?

The Resources section links to key papers like Gentry's thesis, tutorials from Microsoft Research, and courses from KU Leuven, providing both theoretical foundations and practical insights. Community sites like HomomorphicEncryption.org also offer standardization updates and discussions.

Awesome HE — Awesome Homomorphic Encryption Resources

What is Awesome HE?

Awesome Homomorphic Encryption is a curated GitHub repository listing libraries, toolkits, applications, databases, and educational resources related to homomorphic encryption. It helps developers and researchers discover tools for performing computations on encrypted data without needing to decrypt it first, enabling privacy-preserving data analysis and machine learning.

Target Audience

Cryptography researchers, security engineers, data scientists, and software developers working on privacy-sensitive applications who need to explore or implement homomorphic encryption solutions.

Value Proposition

It saves significant research time by aggregating the fragmented HE ecosystem into a single, well-organized list, providing direct links to major libraries like Microsoft SEAL and OpenFHE, practical toolkits, and learning materials.

✨ Awesome - A curated list of amazing Homomorphic Encryption libraries, software and resources

Use Cases

Best For

Finding production-ready FHE libraries for a new project
Learning homomorphic encryption concepts through curated papers and tutorials
Comparing different HE schemes and implementations (BFV, CKKS, TFHE)
Discovering toolkits to compile Python/C++ programs into FHE circuits
Exploring real-world HE applications in databases and machine learning
Staying updated with the latest open-source HE tools and research

Not Ideal For

Teams needing integrated, commercial-grade HE platforms with dedicated technical support and service-level agreements
Beginners seeking interactive, code-along tutorials to learn homomorphic encryption hands-on from scratch
Projects requiring detailed performance benchmarks or comparative analysis to select HE libraries for specific workloads like real-time machine learning inference

Pros & Cons

Pros

Extensive Resource Aggregation

Curates over 50 libraries, toolkits, and applications, including major players like Microsoft SEAL and OpenFHE, saving researchers hours of scattered searching across GitHub and academic papers.

Multi-Language Ecosystem

Features implementations in C++, Rust, Python, Go, and more, as evidenced by libraries like node-seal for JavaScript and TenSEAL for Python tensor operations, catering to diverse development environments.

Practical Application Focus

Showcases real-world uses in databases, ML, and genomics, with direct links to demos like Zama's Hugging Face spaces and OpenMined's PySyft, illustrating HE's potential beyond theory.

Learning Pathway Inclusion

Provides curated links to foundational papers, tutorials, and community sites like FHE.org and Microsoft Research videos, lowering the barrier to entry for newcomers to cryptography.

Cons

No Curated Guidance

As a passive list, it lacks recommendations or comparisons, forcing users to independently evaluate complex trade-offs between schemes like BFV, CKKS, and TFHE without expert insights.

Minimal Hands-On Content

Does not include sample code, step-by-step tutorials, or interactive examples, relying solely on external links that may vary in quality, accessibility, or maintenance over time.

Community-Maintenance Risks

Being community-driven, some links or resources might become outdated or broken, as acknowledged by its open contribution model, requiring users to verify information independently.

Frequently Asked Questions

What is Awesome HE?

Target Audience

Cryptography researchers, security engineers, data scientists, and software developers working on privacy-sensitive applications who need to explore or implement homomorphic encryption solutions.

Value Proposition

Use Cases

Best For

Finding production-ready FHE libraries for a new project
Learning homomorphic encryption concepts through curated papers and tutorials
Comparing different HE schemes and implementations (BFV, CKKS, TFHE)
Discovering toolkits to compile Python/C++ programs into FHE circuits
Exploring real-world HE applications in databases and machine learning
Staying updated with the latest open-source HE tools and research

Not Ideal For

Teams needing integrated, commercial-grade HE platforms with dedicated technical support and service-level agreements
Beginners seeking interactive, code-along tutorials to learn homomorphic encryption hands-on from scratch
Projects requiring detailed performance benchmarks or comparative analysis to select HE libraries for specific workloads like real-time machine learning inference

Pros & Cons

Pros

Extensive Resource Aggregation

Curates over 50 libraries, toolkits, and applications, including major players like Microsoft SEAL and OpenFHE, saving researchers hours of scattered searching across GitHub and academic papers.

Multi-Language Ecosystem

Practical Application Focus

Showcases real-world uses in databases, ML, and genomics, with direct links to demos like Zama's Hugging Face spaces and OpenMined's PySyft, illustrating HE's potential beyond theory.

Learning Pathway Inclusion

Provides curated links to foundational papers, tutorials, and community sites like FHE.org and Microsoft Research videos, lowering the barrier to entry for newcomers to cryptography.

Cons

No Curated Guidance

As a passive list, it lacks recommendations or comparisons, forcing users to independently evaluate complex trade-offs between schemes like BFV, CKKS, and TFHE without expert insights.

Minimal Hands-On Content

Does not include sample code, step-by-step tutorials, or interactive examples, relying solely on external links that may vary in quality, accessibility, or maintenance over time.

Community-Maintenance Risks

Being community-driven, some links or resources might become outdated or broken, as acknowledged by its open contribution model, requiring users to verify information independently.

Frequently Asked Questions

Awesome HE

What is Awesome HE?

Overview

Use Cases

Best For

Not Ideal For

Pros & Cons

Pros

Cons

Frequently Asked Questions

Related Projects

Found a gem we're missing?

Awesome HE

What is Awesome HE?

Overview

Use Cases

Best For

Not Ideal For

Pros & Cons

Pros

Cons

Frequently Asked Questions

Related Projects

Found a gem we're missing?