Zama Product Releases - October 2024

October 8, 2024
  -  
The Zama Team

This month, our team at Zama released a new version of TFHE-rs (v0.8), alongside new versions of Concrete (v2.8) and Concrete ML (v1.7). In November, we will release fhEVM (v0.6), featuring its new fhEVM Coprocessor that enables developers to deploy confidential smart contracts directly on Ethereum. With these releases, Zama continues to build its suite of products to make homomorphic encryption accessible, easy, and fast.

TFHE-rs v0.8

TFHE-rs v0.8 introduces array types and enhances multi-GPU computing. With this release, developers can now work with vectors and tensors more easily. In addition, the enhanced multi-GPU acceleration drastically reduces the computation time for arithmetic operations on GPUs. For instance, multiplying two encrypted 64-bit integers now takes about 100 ms on 8xH100, versus 366 ms on a high-end CPU, bringing a 3.5x speedup.

Read the full release blog post here

Concrete v2.8

Concrete v2.8 introduces two major enhancements: the interoperability between TFHE-rs and Concrete and automatic module tracing for an easier way to fine-tune module compilation. Additionally, we’ve made many improvements to our documentation, including new tutorials showcasing real-world use cases. Concrete v2.8 also includes various optimizations and bug fixes, especially in the Concrete GPU runtime, which makes FHE evaluations faster.

Read the full release blog post here

Concrete ML v1.7

Concrete ML v1.7 introduces several interesting new features. Users can now fine-tune LLMs and neural networks on encrypted data, unlocking more potential privacy-preserving machine learning use cases. Concrete ML v1.7 also adds GPU acceleration that enables faster performance - up to 1-2x speed-up for large neural networks on server-grade GPUs such as the NVIDIA H100. Try out Zama new Hugging Face space demo that showcases a complex use-case: predicting ancestry from encrypted DNA. Finally, Concrete ML now supports Python 3.11 and PyTorch 2.

Read the full release blog post here

fhEVM v0.6

In November, we will release fhEVM v0.6, featuring its new fhEVM Coprocessor that enables developers to deploy confidential smart contracts directly on Ethereum. By leveraging Fully Homomorphic Encryption, developers can seamlessly add confidentiality to Dapps without the need for additional bridging, ensuring end-to-end onchain privacy. This solution is fully compatible with the Ethereum ecosystem, offers a simple user experience, and enhances both security and scalability. Stay tuned!

Register to get notified at launch

Read more related posts

TFHE-rs v0.8: Encrypted Arrays and Improved Multi-GPU Support

TFHE-rs v0.8 introduces array types, enhances multi-GPU computing and many improvements.

Read Article

Concrete v2.8: Interoperability with TFHE-rs and Automatic Module Tracing

Concrete v2.8 introduces the interoperability between TFHE-rs and Concrete, automatic module tracing, and new use-case tutorials.

Read Article

Concrete ML v1.7: Transformer Model Fine-tuning and DNA Ancestry Applications

Concrete ML v1.7 introduces fine-tuning models on encrypted data, GPU acceleration, and DNA ancestry applications.

Read Article