IP Library Granted Patent US 12,652,156
Granted Patent B1
US 12,652,156 · App. 19/294,976 · Granted Jun 9, 2026

System and method for accelerating fully homomorphic encryption computations

Inventors: Oren Vrubel (Tel Aviv, IL); Nimrod Borer (Atlit, IL); Ilan Rosenfeld (Petah Tikva, IL)
Assignee: Chain Reaction, Ltd.
H04L9/0618H04L9/008
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,652,156
App. No.
19/294,976
Granted
Jun 9, 2026
Kind
B1
Abstract

A method and system for fully homomorphic-encryption (FHE) inference is presented. The method includes receiving, by a computing system, an encrypted request comprising a plurality of ciphertexts; loading, in response to the encrypted request, auxiliary data onto a first set of accelerators and a second set of accelerators; loading a first set of the plurality of ciphertexts onto the second set of accelerators; refreshing the first set of ciphertexts by executing, on the first set of accelerators, a bootstrapping operation that uses the auxiliary data; processing the first set of ciphertexts and a second set of the plurality of ciphertexts; and generating a response ciphertext based on the processing of the first and second sets of ciphertexts.

Claims (51)

1 . A method for fully homomorphic-encryption (FHE) inference, the method comprising:

receiving, by a computing system, an encrypted request comprising a plurality of ciphertexts;

loading, in response to the encrypted request, auxiliary data onto a first set of accelerators and a second set of accelerators, wherein, when two or more users share an identical fully-homomorphic-encryption key set, one or more accelerators are assigned as dedicated bootstrapping accelerators for the key set and stores the auxiliary data in on-chip memory for the duration of a session;

loading a first set of the plurality of ciphertexts onto the second set of accelerators;

refreshing the first set of ciphertexts by executing, on the first set of accelerators, a bootstrapping operation that uses the auxiliary data;

processing the first set of ciphertexts and a second set of the plurality of ciphertexts; and

generating a response ciphertext based on the processing of the first and second sets of ciphertexts.

2 . The method of claim 1 , wherein the auxiliary data is session-specific auxiliary data that is loaded into on-chip memory of the first set of accelerators for a duration of a session.

3 . The method of claim 1 , wherein the processing of the first and the second set of ciphertexts comprises performing one or more non-bootstrapping homomorphic operations selected from at least one of vector-matrix multiplication, slot rotation, rescaling, and point-wise activation.

4 . The method of claim 1 , further comprising, prior to the loading of the first set of ciphertexts, determining a batch size at compile time, the batch size defining a number of ciphertexts processed in parallel by the second set of accelerators.

5 . The method of claim 1 , further comprising: transferring plaintext-weight blocks into device memory by direct-memory access while the second set of accelerators is processing ciphertexts.

6 . The method of claim 1 , wherein refreshing the first set of ciphertexts is performed entirely on-chip by the first set of accelerators with no, or reduced access to external memory.

7 . The method of claim 1 , wherein the encrypted request includes prompt-phase ciphertexts, and the first set of accelerators remains dedicated to the encrypted request until a first response ciphertext is generated.

8 . The method of claim 1 , further comprising: storing, in a double-buffer memory associated with each accelerator of the second set, a first plaintext-weight block used for computation while concurrently loading a second plaintext-weight block into an idle portion of the double-buffer memory.

9 . The method of claim 1 , further comprising, after the response ciphertext is generated, releasing at least one accelerator of the first set to process a different encrypted request.

10 . The method of claim 1 , further comprising, during a generation phase corresponding to the encrypted request, assigning a dedicated bootstrapping accelerator to a session that uses a distinct fully-homomorphic-encryption key set, wherein the dedicated bootstrapping accelerator stores the session's auxiliary data in on-chip memory for the duration of the session.

11 . The method of claim 1 , wherein each ciphertext of the first set and the second set encrypts a respective one or more tokens generated by a multi-token large-language-model.

12 . The method of claim 1 , wherein the plurality of ciphertexts represents activations of a convolutional neural network, and the first set of accelerators refreshes the ciphertexts during processing of one or more convolutional layers.

13 . A non-transitory computer-readable medium storing a set of instructions for fully homomorphic-encryption (FHE) inference, the set of instructions comprising:

one or more instructions that, when executed by one or more processing circuitries of a device, cause the device to:

receive, by a computing system, an encrypted request comprising a plurality of ciphertexts;

load, in response to the encrypted request, auxiliary data onto a first set of accelerators and a second set of accelerators, wherein, when two or more users share an identical fully-homomorphic-encryption key set, one or more accelerators are assigned as dedicated bootstrapping accelerators for the key set and stores the auxiliary data in on-chip memory for the duration of a session;

load a first set of the plurality of ciphertexts onto the second set of accelerators;

refresh the first set of ciphertexts by executing, on the first set of accelerators, a bootstrapping operation that uses the auxiliary data;

process the first set of ciphertexts and a second set of the plurality of ciphertexts; and

generate a response ciphertext based on the processing of the first and second sets of ciphertexts.

14 . A system for fully homomorphic-encryption (FHE) inference comprising:

a processing circuitry;

a memory, the memory containing instructions that, when executed by the processing circuitry, configure the system to:

receive, by a computing system, an encrypted request comprising a plurality of ciphertexts;

load, in response to the encrypted request, auxiliary data onto a first set of accelerators and a second set of accelerators, wherein, when two or more users share an identical fully-homomorphic-encryption key set, one or more accelerators are assigned as dedicated bootstrapping accelerators for the key set and stores the auxiliary data in on-chip memory for the duration of a session;

load a first set of the plurality of ciphertexts onto the second set of accelerators;

refresh the first set of ciphertexts by executing, on the first set of accelerators, a bootstrapping operation that uses the auxiliary data;

process the first set of ciphertexts and a second set of the plurality of ciphertexts; and

generate a response ciphertext based on the processing of the first and second sets of ciphertexts.

15 . The system of claim 14 , wherein the auxiliary data is session-specific auxiliary data that is loaded into on-chip memory of the first set of accelerators for a duration of a session.

16 . The system of claim 14 , wherein the processing circuitry, when processing the first and the second set of ciphertexts, is configured to perform one or more non-bootstrapping homomorphic operations selected from at least one of vectormatrix multiplication, slot rotation, rescaling, and point-wise activation.

17 . The system of claim 14 , wherein the memory contains further instructions which when executed by the processing circuitry further configure the system to:

determine a batch size at compile time prior to the loading of the first set of ciphertexts, wherein the batch size is defined by a number of ciphertexts processed in parallel by the second set of accelerators.

18 . The system of claim 14 , wherein the memory contains further instructions which when executed by the processing circuitry further configure the system to:

transfer plaintext-weight blocks into device memory by direct-memory access while the second set of accelerators is processing ciphertexts.

19 . The system of claim 14 , wherein refreshing the first set of ciphertexts is performed entirely on-chip by the first set of accelerators with no, or reduced access to external memory.

20 . The system of claim 14 , wherein the encrypted request includes prompt-phase ciphertexts, and the first set of accelerators remains dedicated to the encrypted request until a first response ciphertext is generated.

21 . The system of claim 14 , wherein the memory contains further instructions which when executed by the processing circuitry further configure the system to:

store, in a double-buffer memory associated with each accelerator of the second set, a first plaintext-weight block used for computation while concurrently loading a second plaintext-weight block into an idle portion of the double-buffer memory.

22 . The system of claim 14 , wherein the memory contains further instructions which when executed by the processing circuitry further configure the system to:

release at least one accelerator of the first set to process a different encrypted request after the response ciphertext is generated.

23 . The system of claim 14 , wherein the memory contains further instructions which when executed by the processing circuitry further configure the system to:

assign a dedicated bootstrapping accelerator to a session that uses a distinct fully-homomorphic-encryption key set during a generation phase corresponding to the encrypted request, wherein the dedicated bootstrapping accelerator stores the session's auxiliary data in on-chip memory for the duration of the session.

24 . The system of claim 14 , wherein each ciphertext of the first set and the second set encrypts a respective one or more tokens generated by a multi-token large-language-model.

25 . The system of claim 14 , wherein the plurality of ciphertexts represents activations of a convolutional neural network, and the first set of accelerators refreshes the ciphertexts during processing of one or more convolutional layers.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 11, 2025
From: VRUBEL, OREN; BORER, NIMROD; ROSENFELD, ILAN
To: CHAIN REACTION, LTD.
Reel/Frame 071984/0948 →
References Cited (35)
US 8515058B1 · Gentry · 2013 [cited by examiner]
US 10075288B1 · Khedr · 2018 [cited by examiner]
US 10541805B2 · Laine · 2020 [cited by examiner]
US 12200101B2 · Agrawal · 2025 [cited by examiner]
US 12368571B1 · Agrawal · 2025 [cited by examiner]
US 20210160225A1 · Liu · 2021 [cited by examiner]
US 20240022393A1 · Park · 2024 [cited by examiner]
US 20240340158A1 · Choi · 2024 [cited by examiner]
US 20240394067A1 · Dimou · 2024 [cited by examiner]
US 20240421971A1 · Agrawal · 2024 [cited by examiner]
US 20250077120A1 · Ohba · 2025 [cited by examiner]
US 20250150254A1 · Adir · 2025 [cited by examiner]
US 20250167976A1 · Van Beirendonck · 2025 [cited by examiner]
US 20250247207A1 · Song · 2025 [cited by examiner]
US 20250259042A1 · Crabtree · 2025 [cited by examiner]
US 20250300806A1 · Vrubel · 2025 [cited by examiner]
US 20250300807A1 · Vrubel · 2025 [cited by examiner]
CN 114201746A · 2022 [cited by examiner]
CN 115776367A · 2023 [cited by examiner]
CN 116048811A · 2023 [cited by examiner]
WO WO2025085808A1 · 2025 [cited by examiner]
H. Nassar, L. Bauer and J. Henkel, “Turbo-FHE: Accelerating Fully Homomorphic Encryption With FPGA and HBM Integration, ” in IEEE Design & Test, vol. 42, No. 3, pp. 86-93, Jun. 2025 (Year: 2025). [cited by examiner]
S. Fan, et al., 2025. FAST: An FHE Accelerator for Scalable-parallelism with Tunable-bit. In Proceedings of the 52nd Annual International Symposium on Computer Architecture (ISCA '25). Association for Computing Machiner… [cited by examiner]
Duong, Phap Ngoc, and Hanho Lee. “Pipelined Key Switching Accelerator Architecture for CKKS-Based Fully Homomorphic Encryption.” Sensors (Basel, Switzerland) vol. 23, 10 4594. May 9, 2023 (Year: 2023). [cited by examiner]
Xiao, Y., Liu, F.-H., Ku, Y.-T., Ho, M.-C., Hsu, C.-F., Chang, M.-C., Hung, S.-H., & Chen, W.-C. (2024). GPU Acceleration for FHEW/TFHE Bootstrapping. IACR Transactions on Cryptographic Hardware and Embedded Systems, 20… [cited by examiner]
R. Agrawal, A. Chandrakasan and A. Joshi, “HEAP: A Fully Homomorphic Encryption Accelerator with Parallelized Bootstrapping,” 2024 ACM/IEEE 51st Annual International Symposium on Computer Architecture (ISCA), Buenos Air… [cited by examiner]
J. Kim et al., “ARK: Fully Homomorphic Encryption Accelerator with Runtime Data Generation and Inter-Operation Key Reuse,” 2022 55th IEEE/ACM International Symposium on Microarchitecture (MICRO), Chicago, IL, USA, 2022,… [cited by examiner]
N.d.). Zama.Ai. Retrieved Aug. 8, 2025, from https://www.zama.ai/products-and-services/privacy-preserving-machine-learning-using-fully-homomorphic-encryption. [cited by applicant]
CryptoLab, & Niobium. (Oct. 10, 2024). CryptoLab and Niobium partner to bring guaranteed privacy to LLM and generative AI applications. Cision PR Newswire. https://www.prnewswire.com/news-releases/cryptolab-and-niobium-… [cited by applicant]
FracTLcore® Compute Fabric. (Jan. 28, 2025). Cornami—Realtime Intelligent Computing for AI, Blockchain, and PPML Workloads with FHE; Fully Homomorphic Encryption, and Plaintext; Cornami, inc. https://cornami.com/fractlc… [cited by applicant]
Ligeti, R. (Oct. 22, 2020). Cornami achieves unprecedented 1,000,000x Acceleration to Deliver Real-Time Fully Homomorphic Encryption (FHE). Cornami—Realtime Intelligent Computing for AI, Blockchain, and PPML Workloads w… [cited by applicant]
Plataniotis□, K. (n.d.). Special issue on privacy and security challenges of generative AI. ITU. Retrieved Aug. 8, 2025, from https://www.itu.int/en/journal/j-fet/2025/003/Pages/default.aspx. [cited by applicant]
Rho, D., Kim, T., Park, M., Kim, J. W., Chae, H., Ryu, E. K., & Cheon, J. H. (2024). Encryption-Friendly LLM Architecture. In arXiv [cs.CR]. http://arxiv.org/abs/2410.02486. [cited by applicant]
A. Kim et al., “General Bootstrapping Approach for RLWE-Based Homomorphic Encryption,” in IEEE Transactions on Computers, vol. 73, No. 1, pp. 86-96, Jan. 2024 (Year: 2024). [cited by applicant]
De Castro, Leo, et al. “EncryptedLLM: Privacy-Preserving Large Language Model Inference via GPU-Accelerated Fully Homomorphic Encryption.” Forty-second International Conference on Machine Leaming. (Year: 2025). [cited by applicant]