Get Chips & Computing every Monday

One free email with what the biggest companies in Chips & Computing filed, were granted and gave up this week.

The Thicket.
Chips & Computing / Week of Nov 16, 2026Get it every Monday →
All Chips & ComputingIssue 001 / IntelThe filing
IntelGrant

Intel secures grants for multi-tile memory management and DNN kernel decomposition.

The filing

On November 10, 2026, Intel Corporation received US grants for technologies including multi-tile memory management and deep neural network acceleration, with the multi-tile patent describing a graphics processor using an interposer to connect chiplets for high-bandwidth memory access (USPTO 12,792,559 (opens in a new tab)) and the DNN patent detailing a 1×N kernel decomposition method for weight generation and activation broadcasting (USPTO 12,792,308 (opens in a new tab)). On November 12, 2026, the company published applications covering maximum exponent-based matrix multiplication, in-field translation of neural processing unit binaries, and CPU-driven silicon initialization offloading (USPTO 20260359005A1 (opens in a new tab)).

This filing’s full text wasn’t read for this issue, so its claims and the applicant’s background aren’t shown. Both are in the record, under Sources.

Expert view

Intel Corporation received US grants on November 10, 2026, for technologies including multi-tile memory management and deep neural network acceleration. The multi-tile memory management patent describes a graphics processor with an interposer connecting a first chiplet containing cache circuitry and a second chiplet with a memory-side cache and memory controller for high-bandwidth memory access (USPTO 12,792,559 (opens in a new tab)). The DNN accelerator patent details a 1×N kernel decomposition method where a weight operand is generated from a filter and input operands are read to internal memories of processing elements for activation broadcasting and MAC operations (USPTO 12,792,308 (opens in a new tab)). Additional grants cover memory module management, Keccak algorithm acceleration, aperture access processors, cluster-wide storage rebuild reduction, RTL bug fixing, NIC interrupt delivery, die-to-die interconnects, scalar core integration, EDA tool output prediction, multi-tile GPUs, phase alignment, privacy-protected attestation, and data center cluster architecture (USPTO 12,792,212 (opens in a new tab)), (USPTO 12,792,345 (opens in a new tab)), (USPTO 12,792,387 (opens in a new tab)), (USPTO 12,792,477 (opens in a new tab)), (USPTO 12,792,493 (opens in a new tab)), (USPTO 12,792,539 (opens in a new tab)), (USPTO 12,792,551 (opens in a new tab)), (USPTO 12,792,560 (opens in a new tab)), (USPTO 12,792,803 (opens in a new tab)), (USPTO 12,793,166 (opens in a new tab)), (USPTO 12,794,662 (opens in a new tab)), (USPTO 12,794,714 (opens in a new tab)), (USPTO 12,794,989 (opens in a new tab)). On November 12, 2026, the company published applications for maximum exponent-based matrix multiplication, in-field translation of neural processing unit binaries, CPU-driven silicon initialization offloading, GPU resource partitioning, configuration access filtering, data center CPU capacity determination, synthetic fixed content memory, tile register pair storage, and efficient sum and signed two's complement values (USPTO 20260359005A1 (opens in a new tab)), (USPTO 20260359026A1 (opens in a new tab)), (USPTO 20260359064A1 (opens in a new tab)), (USPTO 20260359140A1 (opens in a new tab)), (USPTO 20260359158A1 (opens in a new tab)), (USPTO 20260359181A1 (opens in a new tab)), (USPTO 20260359225A1 (opens in a new tab)), (USPTO 20260359381A1 (opens in a new tab)), (USPTO 20260359384A1 (opens in a new tab)).

Sources (24)

Get Chips & Computing every Monday.

© 2026 Fearn AI, Inc. All rights reserved.

Information from the public patent record, not legal advice.

Big Law patents. In plain sight.

The all-in-one patent firm for startups. Former Big Law patent experts and PhDs, on AI that runs on our own GPUs.

Social

Newsletter

Stay updated on the future of intellectual property.

Fearn Legal, LLC is a patent prosecution firm, licensed to file and prosecute patents across the USPTO. The information provided here is for general informational purposes only and does not constitute legal advice. Viewing this information or contacting us through this website does not create an attorney-client or patent agent-client relationship. For personalized legal advice, please reach out to us directly to set up a legal consult or consult directly with a qualified attorney.

Big Law patents. In plain sight.

Social

Newsletter

Stay updated on the future of intellectual property.

Fearn Legal, LLC is a patent prosecution firm, licensed to file and prosecute patents across the USPTO. The information provided here is for general informational purposes only and does not constitute legal advice. Viewing this information or contacting us through this website does not create an attorney-client or patent agent-client relationship. For personalized legal advice, please reach out to us directly to set up a legal consult or consult directly with a qualified attorney.

Big Law patents. In plain sight.

The all-in-one patent firm for startups. Former Big Law patent experts and PhDs, on AI that runs on our own GPUs.

Social

Newsletter

Stay updated on the future of intellectual property.

Fearn Legal, LLC is a patent prosecution firm, licensed to file and prosecute patents across the USPTO. The information provided here is for general informational purposes only and does not constitute legal advice. Viewing this information or contacting us through this website does not create an attorney-client or patent agent-client relationship. For personalized legal advice, please reach out to us directly to set up a legal consult or consult directly with a qualified attorney.