For the Inventor, By the Inventor

Enter assignee name

Mountain View, CA, United States of America

Groq, Inc.

Calculating rank

USPTO Granted Patents = 72

EPO Patents = 5

Trademarks = 1

Forward Citations = 105

Loading Chart...

Loading Chart...

36 inventors (with patents filed for the assignee):

Jonathan Alexander Ross (43 out of 98 patents)

Gregory Michael Thorson (30 out of 76 patents)

Dennis Charles Abts (18 out of 52 patents)

4. John Thompson (11 out of 11 patents)

5. Dinesh Maheshwari (7 out of 73 patents)

6. Christopher Aaron Clark (6 out of 31 patents)

7. Matt Boyd (5 out of 5 patents)

8. Sushma Honnavara-Prasad (5 out of 5 patents)

9. James David Sproch (4 out of 23 patents)

10. Srivathsa Dhruvanarayan (3 out of 20 patents)

11. Tom Hawkins (3 out of 3 patents)

12. Thomas Hawkins (3 out of 4 patents)

13. Andrew Ling (2 out of 16 patents)

14. Michael Bye (2 out of 3 patents)

15. Brian Lee Kurtz (2 out of 3 patents)

Core Competencies:

Tensor Streaming Processor

Power Optimization

Data Transformation

Memory Design

Compiler Operations

Instruction Set Architecture

Deterministic Memory

Energy Provisioning

Processor Architecture

Machine Learning

High-Speed Processors

Clock Waveforms

72 patents: Sort by Forward Citations

1. 12769654 - Deterministic streaming processor architecture having functional units (2026-10-20),

2. 12755732 - Model substitution for efficient development of a solution (2026-10-06),

3. 12754716 - Tiled switch matrix data permutation circuit (2026-10-06),

4. 12688075 - Graph partitioning and implementation of large models on tensor streaming processors (2026-07-21),

5. 12675549 - Error correction in computation (2026-07-07),

6. 12660683 - Tile structures for multiple die tensor streaming processors (2026-06-16),

7. 12561279 - Deterministic memory for tensor streaming processors (2026-02-24),

8. 12547459 - Data transformation algorithms for tensor streaming processor (2026-02-10),

9. 12475363 - Tensor processor visualization and analysis tool (2025-11-18),

10. 12443216 - Generation of clock waveforms based on decoded instructions for more efficient power management in high-speed processors (2025-10-14),

11. 12411762 - Memory design for a processor (2025-09-09),

12. 12373107 - Method to manage periodic dram refresh and maintenance scheduling for predictable dram data access (2025-07-29),

13. 12373018 - Power optimization in an artificial intelligence processor (2025-07-29),

14. 12340300 - Streaming processor architecture (2025-06-24),

15. 12287695 - Energy provisioning (2025-04-29),

Company Overview

Groq, Inc. is a technology company based in Mountain View, California, specializing in advanced computing solutions through its innovative Tensor Streaming Processor architecture. The company focuses on optimizing power consumption and enhancing data transformation capabilities, making significant strides in memory design and compiler operations. Groq's technology is designed to deliver high performance in various applications, emphasizing deterministic memory and energy provisioning to meet the demands of modern computing environments.

Profile summary synthesized from public USPTO records and patent filings.

Company Badge

USPTO Data Jan 1 1976 - Oct 27 2026

72 Patents

#47 in Mountain View, CA

#1,205 in California

#5,110 in US

#10,400 in the World

Mountain View, CA, United States of America

Groq, Inc.

Forward Citations = 105

36 inventors:

Jonathan Alexander Ross (43 out of 98 patents)

Gregory Michael Thorson (30 out of 76 patents)

Dennis Charles Abts (18 out of 52 patents)

4. John Thompson (11 out of 11 patents)

5. Dinesh Maheshwari (7 out of 73 patents)

IDiyas Innovation Intelligence. (2026). Company Patent Portfolio: Groq, Inc.. Retrieved from https://idiyas.com/company/groq

Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile

Data Sources: USPTO Patent Grant XML, Patent Center, EPO & CIPO • Normalized by IDiyas Innovation Graph. Methodology & provenance architecturePlease report any incorrect information to [email protected]

Loading…