Changeflow GovPing Telecom & Technology Pyramid Key-Value Cache Compression for Transfo...
Routine Notice Added Final

Pyramid Key-Value Cache Compression for Transformer Models

Favicon for changeflow.com USPTO Patent Applications - AI & Computing (G06N)
Published
Detected
Email

Summary

USPTO published patent application US20260099695A1 on April 9, 2026, for a method of operating transformer models with algorithmic key-value cache memory allocation across decoding layers. The invention allocates a fixed memory budget progressively across layers, with higher layers receiving smaller cache allocations. Each layer independently determines maximum key-value vector pairs based on its allocated cache.

Published by USPTO on changeflow.com . Detected, standardized, and enriched by GovPing. Review our methodology and editorial standards .

What changed

USPTO published patent application US20260099695A1 for a pyramid key-value cache compression method in transformer models. The application claims a system for allocating a fixed cache memory budget across multiple decoding layers, with progressively higher layers receiving smaller cache allocations. Each layer independently caps the maximum number of key-value vector pairs it retains during token decoding operations.

This patent application affects AI researchers, machine learning engineers, and technology companies developing transformer-based models. If granted, the patent would provide intellectual property protection for cache compression techniques that may be relevant to optimizing large language model inference and deployment. The patent has no compliance deadlines or regulatory obligations associated with its publication.

Archived snapshot

Apr 17, 2026

GovPing captured this document from the original source. If the source has since changed or been removed, this is the text as it existed at that time.

← USPTO Patent Applications

PYRAMID KEY-VALUE CACHE COMPRESSION FOR TRANSFORMER MODELS

Application US20260099695A1 Kind: A1 Apr 09, 2026

Inventors

Wen XIAO, Wei XIONG, Abedelkader ASI, Zefan CAI

Abstract

A method for operating a transformer model includes algorithmically allocating a fixed budget for a key-value cache between multiple decoding layers per an allocation scheme that ensures progressively higher decoding layers in the transformer model are allocated progressively smaller quantities of cache memory. The method further includes configuring each of the multiple decoding layers of the transformer model to retain no more than a maximum number of key-value vector pairs in the key-value cache during a token decoding operation, the maximum number of key-value vector pairs being independently determined for each decoding layer of the multiple decoding layers based on the cache memory that is allocated to the decoding layer.

CPC Classifications

G06N 3/045

Filing Date

2024-10-09

Application No.

18910974

View original document →

Get daily alerts for USPTO Patent Applications - AI & Computing (G06N)

Daily digest delivered to your inbox.

Free. Unsubscribe anytime.

About this page

What is GovPing?

Every important government, regulator, and court update from around the world. One place. Real-time. Free. Our mission

What's from the agency?

Source document text, dates, docket IDs, and authority are extracted directly from USPTO.

What's AI-generated?

The summary, classification, recommended actions, deadlines, and penalty information are AI-generated from the original text and may contain errors. Always verify against the source document.

Last updated

Classification

Agency
USPTO
Published
April 9th, 2026
Instrument
Notice
Legal weight
Non-binding
Stage
Final
Change scope
Minor
Document ID
US20260099695A1
Docket
18910974

Who this affects

Applies to
Technology companies Manufacturers
Industry sector
5112 Software & Technology
Activity scope
Patent filing AI model optimization
Geographic scope
United States US

Taxonomy

Primary area
Intellectual Property
Operational domain
Legal
Topics
Artificial Intelligence Software & Technology

Get alerts for this source

We'll email you when USPTO Patent Applications - AI & Computing (G06N) publishes new changes.

Free. Unsubscribe anytime.

You're subscribed!