Blog
 / 
Announcements
TFHE-rs

TFHE-rs v0.11: Strings, Faster Zero Knowledge Proof, Encrypted Arrays on GPU

January 14, 2025
  -  
Jean-Baptiste Orfila, Arthur Meyre, Agnes Leroy

TFHE-rs v0.11 brings several major improvements and new features. This release introduces significantly enhanced performance for Zero Knowledge Proof and a new module for encrypted strings usable in the High Level API via the [.c-inline-code]FheAsciiString[.c-inline-code] type. Besides, default cryptographic parameters now follow a tweaked uniform (TUniform) noise distribution instead of a Gaussian one. GPU performance has also been improved: the 64-bit addition is 30% faster compared to v0.8. Finally, it is now possible to perform computations on arrays of ciphertexts easily on GPU.

Zero Knowledge Proof v2

The new Zero Knowledge proof construction is now faster both for generating and verifying a proof, as detailed in the benchmark table below.

You can follow along the documentation guide to use Zero Knowledge Proofs in the High-Level API.

FHE strings

One of the highlights of this release is the FHE strings module, based on an open source contribution from user JoseSK999 during the Zama Bounty Program Season 5. It is also available in the High-Level API via the [.c-inline-code]FheAsciiString[.c-inline-code] type. FHE strings support the same APIs as Rust’s str type. You can find the guide on FHE strings and how to use them in the documentation.

Here is a preview of what you can do with the new primitive with the integer and strings features enabled:

use tfhe::prelude::*;
use tfhe::{generate_keys, set_server_key, ClearString, ConfigBuilder, FheAsciiString};

fn main() {
    let config = ConfigBuilder::default().build();
    let (cks, sks) = generate_keys(config);

    set_server_key(sks);

    // Encrypt without padding, does not hide the string length and has better performance
    let string = FheAsciiString::try_encrypt("TFHE-rs rocks!", &cks).unwrap();

    // Encrypt with padding, hide the true length of the string
    let search_string = FheAsciiString::try_encrypt_with_padding("is meh", 1, &cks).unwrap();

    // We can also use clear strings
    let clear_search_string = ClearString::new("rocks".to_string());

    // Does our initial string contain "is meh"?
    let does_not_contain = string.contains(&search_string);

    // Does our initial string contain "rocks"?
    let contains = string.contains(&clear_search_string);

    // Decrypt
    let decrypted_does_not_contain = does_not_contain.decrypt(&cks);
    let decrypted_contains = contains.decrypt(&cks);

    // Check all worked properly
    assert!(!decrypted_does_not_contain);
    assert!(decrypted_contains);
}

We expect most string processing to consist mostly of comparison and contains operation, here are the current benchmarks on an hpc7a.96xlarge AWS machine:

Computing over encrypted arrays on GPU

TFHE-rs v0.11 introduces n-dimensional arrays (or tensors) for encrypted data on GPU. The supported operations are the same as on CPU, including the comparison between vectors and the search in a vector of encrypted data.

The following example displays how to use array types on GPU: it shows how to extract submatrices of size 2x2 from two 4x4 matrices, perform addition, and then add a clear matrix to the previous results. You can find more details on how to use homomorphic arrays in the documentation.

use tfhe::{ConfigBuilder, set_server_key, CpuFheUint32Array, ClearArray, ClientKey, CompressedServerKey};
use tfhe::array::GpuFheUint32Array;
use tfhe::prelude::*;

fn main() {
    let config = ConfigBuilder::default().build();
    let cks = ClientKey::generate(config);
    let compressed_server_key = CompressedServerKey::new(&cks);

    let gpu_key = compressed_server_key.decompress_to_gpu();

    set_server_key(gpu_key);

    let num_elems = 4 * 4;
    let clear_xs = (0..num_elems as u32).collect::<Vec<_>>();
    let clear_ys = vec![1u32; num_elems];

    // Encrypted 2D array with values
    // [[  0,  1,  2,  3]
    //  [  4,  5,  6,  7]
    //  [  8,  9, 10, 11]
    //  [ 12, 13, 14, 15]]
    // and shape 4x4
    let xs = GpuFheUint32Array::try_encrypt((clear_xs.as_slice(), vec![4, 4]), &cks).unwrap();
    // Encrypted 2D array with values
    // [[  1,  1,  1,  1]
    //  [  1,  1,  1,  1]
    //  [  1,  1,  1,  1]
    //  [  1,  1,  1,  1]]
    // and shape 4x4
    let ys = GpuFheUint32Array::try_encrypt((clear_ys.as_slice(), vec![4, 4]), &cks).unwrap();

    assert_eq!(xs.num_dim(), 2);
    assert_eq!(xs.shape(), &[4, 4]);
    assert_eq!(ys.num_dim(), 2);
    assert_eq!(ys.shape(), &[4, 4]);

    // Take a sub slice
    //  [[ 10, 11]
    //   [ 14, 15]]
    let xss = xs.slice(&[2..4, 2..4]);
    // Take a sub slice
    //  [[  1,  1]
    //   [  1,  1]]
    let yss = ys.slice(&[2..4, 2..4]);

    assert_eq!(xss.num_dim(), 2);
    assert_eq!(xss.shape(), &[2, 2]);
    assert_eq!(yss.num_dim(), 2);
    assert_eq!(yss.shape(), &[2, 2]);

    let r = &xss + &yss;

    // Result is
    //  [[ 11, 12]
    //   [ 15, 16]]
    let result: Vec = r.decrypt(&cks);
    assert_eq!(result, vec![11, 12, 15, 16]);

    // Clear 2D array with values
    //  [[  10,  20]
    //   [  30,  40]]
    let clear_array = ClearArray::new(vec![10u32, 20u32, 30u32, 40u32], vec![2, 2]);
}

Improved GPU performance

This version brings performance improvements to the GPU backend. The 64-bit addition is 30% faster than v0.8. The Figure 1 below shows the current performance on CPU & GPU for the 64-bit addition and multiplication.

The optimal number of GPUs per operation varies depending on the operation itself and the integer precision specified by the user. Comprehensive arrays of benchmark results for both single and multiple GPUs across all specified precisions are available in the documentation.

Additional features and improvements

TFHE-rs v0.11 introduces several other features or improvements, including:

  • New GPU operations: Support for absolute value and division on homomorphic signed integers.
  • Faster CPU operations: Encrypted shifts/rotations on the CPU are now 28% faster.
  • Parameter updates: All default cryptographic parameters have been updated to use a new noise distribution, called TUniform. The main advantage of this distribution is to be bounded, whereas the usual Gaussian one is not. In some practical cases, this can simplify the use of homomorphic computation. The mathematical definition can be found in this documentation.
  • Code stability: This release is focused on improving the overall code stability. The complete list of novelties is available in the release note.

The next release of TFHE-rs will continue to improve performance, features and stability! Stay tuned for the upcoming update!

Additional links

Read more related posts

No items found.
Libraries
Concrete ↗ Concrete ML ↗FHEVM ↗TFHE-rs ↗
Products & Services
Privacy-preserving Machine learningConfidential blockchainThreshold key management SYSTEM
Developers
BlogDOCUMENTATION ↗‍GITHUB ↗ FHE resources ↗ Research papers ↗Bounty Program ↗FHE STATE OS
Company
Aboutintroduction to fheEventsMediaCareers ↗Legal
Contact
Talk to an expertCONTACT USXDiscordTelegramALL community channels
Privacy is necessary for an open society in the electronic age. Privacy is not secrecy. A private matter is something one doesn't want the whole world to know, but a secret matter is something one doesn't want anybody to know. Privacy is the power to selectively reveal oneself to the world.If two parties have some sort of dealings, then each has a memory of their interaction. Each party can speak about their own memory of this; how could anyone prevent it? One could pass laws against it, but the freedom of speech, even more than privacy, is fundamental to an open society; we seek not to restrict any speech at all. If many parties speak together in the same forum, each can speak to all the others and aggregate together knowledge about individuals and other parties. The power of electronic communications has enabled such group speech, and it will not go away merely because we might want it to.Since we desire privacy, we must ensure that each party to a transaction have knowledge only of that which is directly necessary for that transaction. Since any information can be spoken of, we must ensure that we reveal as little as possible. In most cases personal identity is not salient. When I purchase a magazine at a store and hand cash to the clerk, there is no need to know who I am. When I ask my electronic mail provider to send and receive messages, my provider need not know to whom I am speaking or what I am saying or what others are saying to me; my provider only need know how to get the message there and how much I owe them in fees. When my identity is revealed by the underlying mechanism of the transaction, I have no privacy. I cannot here selectively reveal myself; I must always reveal myself.Therefore, privacy in an open society requires anonymous transaction systems. Until now, cash has been the primary such system. An anonymous transaction system is not a secret transaction system. An anonymous system empowers individuals to reveal their identity when desired and only when desired; this is the essence of privacy.Privacy in an open society also requires cryptography. If I say something, I want it heard only by those for whom I intend it. If the content of my speech is available to the world, I have no privacy. To encrypt is to indicate the desire for privacy, and to encrypt with weak cryptography is to indicate not too much desire for privacy. Furthermore, to reveal one's identity with assurance when the default is anonymity requires the cryptographic signature.We cannot expect governments, corporations, or other large, faceless organizations to grant us privacy out of their beneficence. It is to their advantage to speak of us, and we should expect that they will speak. To try to prevent their speech is to fight against the realities of information. Information does not just want to be free, it longs to be free. Information expands to fill the available storage space. Information is Rumor's younger, stronger cousin; Information is fleeter of foot, has more eyes, knows more, and understands less than Rumor.We must defend our own privacy if we expect to have any. We must come together and create systems which allow anonymous transactions to take place. People have been defending their own privacy for centuries with whispers, darkness, envelopes, closed doors, secret handshakes, and couriers. The technologies of the past did not allow for strong privacy, but electronic technologies do.We the Cypherpunks are dedicated to building anonymous systems. We are defending our privacy with cryptography, with anonymous mail forwarding systems, with digital signatures, and with electronic money.Cypherpunks write code. We know that someone has to write software to defend privacy, and since we can't get privacy unless we all do, we're going to write it. We publish our code so that our fellow Cypherpunks may practice and play with it. Our code is free for all to use, worldwide. We don't much care if you don't approve of the software we write. We know that software can't be destroyed and that a widely dispersed system can't be shut down.Cypherpunks deplore regulations on cryptography, for encryption is fundamentally a private act. The act of encryption, in fact, removes information from the public realm. Even laws against cryptography reach only so far as a nation's border and the arm of its violence. Cryptography will ineluctably spread over the whole globe, and with it the anonymous transactions systems that it makes possible.For privacy to be widespread it must be part of a social contract. People must come and together deploy these systems for the common good. Privacy only extends so far as the cooperation of one's fellows in society. We the Cypherpunks seek your questions and your concerns and hope we may engage you so that we do not deceive ourselves. We will not, however, be moved out of our course because some may disagree with our goals.The Cypherpunks are actively engaged in making the networks safer for privacy. Let us proceed together apace.Onward. By Eric Hughes. 9 March 1993.