Patentlyze company tracker

New AMD Patents

AMD's patents span Ryzen and EPYC CPUs, Radeon graphics, Instinct AI accelerators, chiplet packaging, and the Xilinx adaptive computing portfolio.

Through the USPTO’s September 24, 2026 release, Patentlyze has covered 99 notable AMD patent applications, including 16 published in the last 30 days.

99

total filings tracked

16

published last 30 days

38

published last 90 days

This week's published patents

Week of September 24, 2026 · 10 filings covered · refreshes every week

AMD's 10 newest filings sort into three clusters:

Multi-chip memory sharing. Two filings cover interconnected processor layouts where chips share memory with neighbors, including a shared memory block that sits between AI processor groups and a grid where adjacent chips access each other's memory directly.
GPU efficiency and scheduling. Four filings target how GPUs manage their own workload, covering self-directed math scheduling without CPU involvement, keeping GPUs busy while memory catches up, trimming memory waste in ray-traced games, and redirecting unused storage toward AI processing.
Graphics and game image quality. The remaining four filings address visual output, including AI-driven texture compression, feeding AI training examples to sharpen game images, drawing less detail during camera movement, and displaying visuals before a PC finishes booting.

What AMD is working on

Based on AMD's activity across the watchlists we track · tracked programs only, not all of AMD's output

Newly emerging

what qualifies: first AMD filing within the last 8 weeks

The AI guardrails race	1 filing, Aug 14
AI agents that act for you	1 filing, Aug 14

Contested races

what qualifies: 3 or more companies in the same watchlist, AMD among them

The AI chip wars	39 of 309 · 12 companies
The GPU rendering race	16 of 109 · 10 companies

AMD patent questions, answered

Numbers pull live from the database, so answers never go stale

What is AMD working on right now?

AMD's tracked filings sit inside contested races, including The AI chip wars and The GPU rendering race. The breakdown above shows each program’s filing count and latest activity.

How many patents did AMD publish this month?

0 AMD patent applications have published in October 2026 so far, after 16 in September. Patentlyze has covered 99 AMD filings since tracking began in May 2026.

Does an AMD patent mean the feature will launch?

No. A published application is not a product announcement. Companies file to protect options, and most patented ideas never ship in the form described. What filings do show is where AMD is spending research effort right now.

Where can I find AMD’s newest patents?

This page updates every week, when the USPTO publishes its new batch of applications. The newest AMD filings appear in the list below.

Top tech areas

Top inventors

Juan J. NOGUERA SERRA 7
Mr. Francisco BARAT QUESADA 6
Stuart David Simpson Biles 5
Michael Estlick 5
David Patrick CLARKE 4

Want this weekly breakdown for a company we don't cover? Patentlyze Pro →

Recent highlights

AMD Patents a Shared Memory Block That Sits Between AI Processor Groups

September 24, 2026

AMD Files Patent for a Processor Grid Where Chips Share Memory With Their Neighbors

September 24, 2026

AMD Patents a GPU That Schedules Its Own Math Work Without Asking the CPU

September 24, 2026

All AMD filings

Chip Designer AMD Patents an AI Processor That Runs Two Types of Math at Once

AMD Patents Technology That Keeps Its Chips Working Without Waiting for Instructions

Xilinx Patents a Way to Trace AI Model Bugs Back to Their Source on Neural Chips

New AMD Patent Locks AI Models Inside Chips Against Software Theft

AMD Patents Technology to Block Protected Video on Displays Failing Encryption Standards

AMD Patents a Processor Trick That Cuts Steps From High-Volume Math

ATI Technologies Patents a Chip Power System That Adjusts Its Own Safety Buffer

AMD Patents a Chip That Ignores AI Model Zeros to Cut Memory Use

AMD Patents an AI That Tunes Your PC Settings and Rolls Back Bad Ones

Xilinx Patents a Way to Fill Idle Chip Space During AI Calculations

AMD Patents a Faster Way for Chips to Handle AI's Favorite Number Format

AMD Patents a Chip That Runs Two AI Systems in Sync to Score Similarities

AMD's New Patent Lets Your PC Tell You Exactly What It Needs to Run Faster

Xilinx Patents a Method to Make AI Models Smaller Without Losing Accuracy

AMD Patent Stores AI Model Data Efficiently to Speed Up Processing

AMD Patent Reduces Hardware Storage Costs by Generating AI Memory Data On Demand

ATI Technologies Patents a Way to Run Motion-Blur Calculations in Reverse

AMD Patents an AI Malware Detector That Runs Outside Your Main Processor

Dedicated AI Compression Chips Emerge From New AMD Patent

AMD Files Patent to Accelerate In-Game Lighting Through Neural Networks

AMD Patents a Hardware Shortcut for the Number-Conversion Work AI Chips Do Constantly

AMD Patent Uses Boolean Matrix Factorization to Shrink AI Models for FPGA Deployment

ATI Technologies Patents a Display System That Dims Each Screen Zone Separately

AMD Patents a Battery-Saving Way to Hand Audio Processing Off to a Dedicated Chip

AMD Patents Software That Reads Your Messages and Automatically Handles Business Tasks

A New Patent Lets Devices Automatically Find and Pull AI Models From the Network

AMD Patents a System That Alerts Developers When Programs Access Forbidden Memory

AMD Patents a Faster Way to Wrap 3D Objects for Ray Tracing

ATI Technologies Patents an AI That Picks Its Own Waypoints for Game Characters

AMD Patent Reveals How Borrowed Math Hardware Could Cut Ray Tracing Costs

AMD Patents a Chip That Picks Its Own Math Engine Based on AI Number Formats

AMD Patents Software That Teaches Computers to Run Programs in Smarter Order

AMD Patents an Image-Resizing System That Adjusts Its Method Based on Visual Complexity

ATI Technologies Patents a Fix for Misaligned Isometric Game Images

ATI Technologies Patents a Way to Cut Wasted Ray Tracing Tests in 3D Scenes

AMD Patents a System That Auto-Compresses Data When AI Chips Share Too Much at Once