# US 2026/0064803 A1 · Matrix multiply-accumulate accelerators for mma operations

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0064803 A1
- Document ID: US20260064803A1
- Status: Published application
- Kind: A1
- Application number: 18/819,979
- Filed: 2024-08-29
- Published: 2026-03-05
- Earliest claimed: 2024-08-29
- Assignees: NVIDIA Corporation
- Applicants: NVIDIA Corporation
- Inventors: Harold Carter Edwards; Vijay Harshad Thakkar; Gokul Ramaswamy Hirisave Chandra Shekhara; Edward H. Gornish; Rishkul Kulkarni; Maciej Piotr Tyrlik; Sean Jeffrey Treichler; Chao Li
- Web page: https://patentgrep.com/patent/US20260064803A1

## CPC

- G06F 9/3001: {Arithmetic instructions}
- G06F 17/16: Matrix or vector computation {, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization (matrix transposition G06F7/78)}
- G06F 9/3887: {controlled by a single instruction for multiple data lanes [SIMD]}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Apparatuses, systems, and techniques to perform an matrix multiply accumulate (MMA) instruction to cause a plurality of portions of an MMA operation to be performed using a corresponding plurality of MMA accelerators. In at least one embodiment, a processor retrieves a plurality of matrix information from a memory that exclusively stores and performs a multiplication computation using said matrix information.

## Claims

### Claim 1

A processor comprising: one or more circuits to perform an matrix multiply accumulate (MMA) instruction to cause a plurality of portions of an MMA operation to be performed using a corresponding plurality of MMA accelerators.

### Claim 2 (depends on 1)

The processor of claim 1, wherein a first operand of the MMA instruction is to be exclusively stored in a tensor memory and a second operand of the MMA instruction is to be stored in a shared memory.

### Claim 3 (depends on 1)

The processor of claim 1, wherein information used by the MMA operation is to be exclusively stored in response to the MMA instruction.

### Claim 4 (depends on 1)

The processor of claim 1, wherein the one or more circuits further perform one or more second instructions concurrently with the MMA operation.

### Claim 5 (depends on 1)

The processor of claim 1, wherein the plurality of portions are to be exclusively stored in a plurality of shared memories and are each accessible by the plurality of MMA accelerators.

### Claim 6 (depends on 1)

The processor of claim 1, wherein the MMA instruction further comprises metadata that identifies an operating mode of the plurality of MMA accelerators.

### Claim 7 (depends on 1)

The processor of claim 1, wherein a single thread of a cooperative thread array performs the instruction that causes the plurality of MMA accelerators to perform the MMA operation.

### Claim 8

A system comprising: one or more processors having one or more circuits to perform an matrix multiply accumulate (MMA) instruction to cause a plurality of portions of an MMA operation to be performed using a corresponding plurality of MMA accelerators.

### Claim 9 (depends on 8)

The system of claim 8, wherein the one or more processors are further to perform the MMA operation in response to the MMA instruction concurrently with one or more second operations.

### Claim 10 (depends on 8)

The system of claim 8, wherein a first operand of the MMA instruction is to be exclusively stored in a tensor memory and a second operand of the MMA instruction is to be stored in a shared memory.

### Claim 11 (depends on 8)

The system of claim 8, wherein information used by the MMA operation is to be exclusively stored in response to the MMA instruction.

### Claim 12 (depends on 8)

The system of claim 8, wherein one or more processors exclusively store the plurality of portions in a plurality of shared memories that are each accessible by the plurality of MMA accelerators.

### Claim 13 (depends on 8)

The system of claim 8, wherein the MMA operation comprises performing matrix multiplication computations in response to the MMA instruction and accumulating results of the computations in an exclusive storage.

### Claim 14 (depends on 8)

The system of claim 8, wherein the plurality of portions correspond to a plurality of operands of the MMA operation.

### Claim 15

A method comprising: performing an matrix multiply accumulate (MMA) instruction to cause a plurality of portions of an MMA operation to be performed using a corresponding plurality of MMA accelerators.

### Claim 16 (depends on 15)

The method of claim 15, further comprising: exclusively storing a first operand of the MMA instruction in a tensor memory and a second operand of the MMA instruction in a shared memory.

### Claim 17 (depends on 15)

The method of claim 15, further comprising: performing one or more second instructions concurrently with the MMA operation.

### Claim 18 (depends on 15)

The method of claim 15, further comprising: using a single thread of a cooperative thread array to perform the MMA instruction that causes the plurality of MMA accelerators to perform the MMA operation.

### Claim 19 (depends on 15)

The method of claim 15, wherein the plurality of portions correspond to a plurality of operands of the MMA operation.

### Claim 20 (depends on 15)

The method of claim 15, wherein a result of the MMA operation is to be exclusively stored in a memory.
