# US 2026/0228001 A1 · Apparatus and method for processing 4-bit floating-point inputs using scaled 5-bit integers for artificial intelligence (ai) workloads

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

- Number: US 2026/0228001 A1
- Document ID: US20260228001A1
- Status: Published application
- Kind: A1
- Application number: 19/632,160
- Filed: 2026-03-27
- Published: 2026-08-06
- Earliest claimed: 2026-03-27
- Assignees: Intel Corporation
- Applicants: Intel Corporation
- Inventors: Chuanjun Zhang; Matthew Adiletta; Simon C. Steely, Jr.
- Web page: https://patentgrep.com/patent/US20260228001A1

## CPC

- G06F 9/3001: {Arithmetic instructions}
- G06F 9/3013: {according to data content, e.g. floating-point registers, address registers}
- G06F 9/30145: {Instruction analysis, e.g. decoding, instruction word fields}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Apparatus and method for scaled integer 5-bit based FP 4 processing. For example, a method comprises: loading a plurality of source 4-bit floating-point (FP4) data elements in one or more registers; determining pairs of INT5 scalar values, INT5 bases, and INT5 signs corresponding to pairs of the source FP4 data elements; performing decoding of each pair of INT5 scalar values in accordance with a map comprising a reduced set of logic gates configured to map each INT5 scalar value directly to generate a corresponding INT5 scalar result; multiplying each corresponding pair of INT5 bases circuitry to generate a base product; generating a two's complement representation of the base product, the two's complement representation comprising a plurality of result bits; and selecting, by a multiplexor network, a set of the result bits based on the corresponding INT5 scalar result, the set of the result bits comprising a dot-product result.

## Claims

### Claim 1

A processor, comprising:
one or more registers to store a plurality of source 4-bit floating-point (FP4) data elements; and
execution circuitry to execute an instruction having fields to indicate a plurality of pairs of source FP4 data elements on which to perform dot product operations, the execution circuitry to determine pairs of INT5 scalar values, INT5 bases, and INT5 signs corresponding to the pairs of FP4 data elements, the execution circuitry comprising:
decoding circuitry to perform decoding of each pair of INT5 scalar values in accordance with a map comprising a reduced set of logic gates configured to map each INT5 scalar value directly to generate a corresponding INT5 scalar result;
a multiplier to multiply each corresponding pair of INT5 bases circuitry to generate a base product;
two's complement circuitry to generate a two's complement representation of the base product, the two's complement representation comprising a plurality of result bits; and
a multiplexor network to select a set of the result bits based on the corresponding INT5 scalar result, the set of the result bits comprising a dot-product result.

### Claim 2 (depends on 1)

The processor of claim 1, wherein the execution circuitry includes circuitry to convert the plurality of pairs of source FP4 data elements to corresponding pairs of INT5 scalar values, INT5 bases, and INT5 signs.

### Claim 3 (depends on 2)

The processor of claim 2, wherein the INT5 scalar values comprise 2-bit scalar values, the INT5 bases comprise 2-bit base values, and the INT5 sign values comprise 1-bit sign values.

### Claim 4 (depends on 3)

The processor of claim 3, wherein the reduced set of logic gates comprise a first set of logic gates associated with an INT5 scalar result of 1, a second set of logic gates associated with an INT5 scalar result of 2, a third set of logic gates associated with an INT5 scalar result of 4, a fourth set of logic gates associated with an INT5 scalar result of 8, and a fifth set of logic gates associated with an INT5 scalar result of 16.

### Claim 5 (depends on 4)

The processor of claim 4, wherein the multiplexor network comprises a plurality of multiplexors, each multiplexor configured to select a corresponding result bit of the set of result bits based on the corresponding INT5 scalar result.

### Claim 6 (depends on 5)

The processor of claim 5, wherein the plurality of multiplexors include multiplexors having different multiplexor sizes, wherein an n:1 multiplexor size is configured for each bit position of the set of result bits, where n is a number of potential bit sources from which a corresponding bit can be selected for the bit position.

### Claim 7 (depends on 1)

The processor of claim 1, wherein the reduced set of logic gates comprises a minimized or optimized set of logic gates generated by a Karnaugh map (K-map).

### Claim 8 (depends on 7)

The processor of claim 7, wherein the K-map comprises a plurality of data structures, each data structure corresponding to one of a plurality of potential INT5 scalar results and indicating corresponding pairs of INT5 scalar values.

### Claim 9

A non-transitory machine-readable medium having program code stored thereon which, when executed by one or more processors, causes the one or more processors to perform operations, comprising:
loading a plurality of source 4-bit floating-point (FP4) data elements in one or more registers;
determining pairs of INT5 scalar values, INT5 bases, and INT5 signs corresponding to pairs of the source FP4 data elements;
performing decoding of each pair of INT5 scalar values in accordance with a map comprising a reduced set of logic gates configured to map each INT5 scalar value directly to generate a corresponding INT5 scalar result;
multiplying each corresponding pair of INT5 bases circuitry to generate a base product;
generating a two's complement representation of the base product, the two's complement representation comprising a plurality of result bits; and
selecting, by a multiplexor network, a set of the result bits based on the corresponding INT5 scalar result, the set of the result bits comprising a dot-product result.

### Claim 10 (depends on 9)

The machine-readable medium of claim 9, further comprising program code to cause the one or more processors to perform the operations of:
converting the plurality of pairs of source FP4 data elements to corresponding pairs of INT5 scalar values, INT5 bases, and INT5 signs.

### Claim 11 (depends on 10)

The machine-readable medium of claim 10, wherein the INT5 scalar values comprise 2-bit scalar values, the INT5 bases comprise 2-bit base values, and the INT5 sign values comprise 1-bit sign values.

### Claim 12 (depends on 11)

The machine-readable medium of claim 11, wherein the reduced set of logic gates comprise a first set of logic gates associated with an INT5 scalar result of 1, a second set of logic gates associated with an INT5 scalar result of 2, a third set of logic gates associated with an INT5 scalar result of 4, a fourth set of logic gates associated with an INT5 scalar result of 8, and a fifth set of logic gates associated with an INT5 scalar result of 16.

### Claim 13 (depends on 12)

The machine-readable medium of claim 12, wherein the multiplexor network comprises a plurality of multiplexors, each multiplexor configured to select a corresponding result bit of the set of result bits based on the corresponding INT5 scalar result.

### Claim 14 (depends on 13)

The machine-readable medium of claim 13, wherein the plurality of multiplexors include multiplexors having different multiplexor sizes, wherein an n:1 multiplexor size is configured for each bit position of the set of result bits, where n is a number of potential bit sources from which a corresponding bit can be selected for the bit position.

### Claim 15 (depends on 9)

The processor of claim 9, wherein the reduced set of logic gates comprises a minimized or optimized set of logic gates generated by a Karnaugh map (K-map).

### Claim 16 (depends on 9)

The machine-readable medium of claim 9, wherein the K-map comprises a plurality of data structures, each data structure corresponding to one of a plurality of potential INT5 scalar results and indicating corresponding pairs of INT5 scalar values.

### Claim 17

A method, comprising:
loading a plurality of source 4-bit floating-point (FP4) data elements in one or more registers;
determining pairs of INT5 scalar values, INT5 bases, and INT5 signs corresponding to pairs of the source FP4 data elements;
performing decoding of each pair of INT5 scalar values in accordance with a map comprising a reduced set of logic gates configured to map each INT5 scalar value directly to generate a corresponding INT5 scalar result;
multiplying each corresponding pair of INT5 bases circuitry to generate a base product;
generating a two's complement representation of the base product, the two's complement representation comprising a plurality of result bits; and
selecting, by a multiplexor network, a set of the result bits based on the corresponding INT5 scalar result, the set of the result bits comprising a dot-product result.

### Claim 18 (depends on 17)

The method of claim 17, further comprising:
converting the plurality of pairs of source FP4 data elements to corresponding pairs of INT5 scalar values, INT5 bases, and INT5 signs.

### Claim 19 (depends on 18)

The method of claim 18, wherein the INT5 scalar values comprise 2-bit scalar values, the INT5 bases comprise 2-bit base values, and the INT5 sign values comprise 1-bit sign values.

### Claim 20 (depends on 18)

The method of claim 18, wherein the K-map comprises a plurality of data structures, each data structure corresponding to one of a plurality of potential INT5 scalar results and indicating corresponding pairs of INT5 scalar values.
