Book a demo

Die-to-Die Interface Patents: Who Leads, Where Filings Cooled 2026

Die-to-Die Interface Patents: Who Leads, Where Filings Cooled 2026
https://www.patsnap.com/resources/blog/rd-blog/die-to-die-interface-energy-efficiency-patent-landscape/ · Patsnap · data cut-off 2026-07-31 · downloaded from the live page
Patent Landscape · Computing Architecture
Die-to-Die Interface Energy Efficiency Patents: Filing Trends and Where the Field Stands
  • Filings peaked in 2022 at 1,022 and have since declined, with the most recent full year sitting well below that mark — a maturing filing cycle rather than a growing one.
  • The ranked leaders hold only 8.4% of all 114,847 records between the top five, so no single assignee controls the field; claim space is fragmented across a long tail.
  • Momentum among the largest filers is falling sharply year-on-year, with drops of 79% to 97% among the most active named assignees in the latest year alone.
Get a prior-art report on your approach
114.8K
Published Records
8%
Top-5 Share of All Records
-22%
Filing Growth 2021→2024
US
Leading Jurisdiction

Filing growth compares 2021 (925 records) with 2024 (719) — a three-year span. 2024 is the most recent year we treat as complete: publication lags filing by roughly 18 months, so 2025 onwards are still filling in and any growth rate that ends there would understate the field. Top-5 share is the combined record count of the five largest assignees divided by all 114,847 records in scope (CR5), not by the ranked leaders only.

Published byPatsnap Research··7 min readSourced from Patsnap Eureka
Overview

What the die-to-die interface energy efficiency filing record shows

Die-to-die interface energy efficiency covers the circuits, protocols and packaging choices that let separate silicon dies exchange data without the power cost of a traditional monolithic bus — clock forwarding schemes, lane repair logic, and the packaging techniques that keep signal reach short. The dataset behind this page spans 114,847 published records filed between 2015 and mid-2026, drawn from a search string built around die and chiplet interconnect PHY terms combined with energy, latency-budget and protocol-overhead qualifiers.

Because publication typically lags filing by around eighteen months, the most recent year in any trend line understates real filing activity — 2026's count is a partial year, not a collapse. Read the trend as a multi-year arc rather than a single data point.

Filing activity, 2017–2026
  1. 1INTEL CORP3,740
  2. 2MICRON TECHNOLOGY INC2,168
  3. 3TAIWAN SEMICONDUCTOR MANUFACTURING CO LTD1,924
  4. 4QUALCOMM INC966
  5. 5TEXAS INSTRUMENTS INC852
  6. 6LG ENERGY SOLUTION LTD804
  7. 7HEWLETT PACKARD DEVELOPMENT COMPANY LP733
  8. 8SAMSUNG ELECTRONICS CO LTD605
  9. 9TOYOTA JIDOSHA KK538
  10. 103M INNOVATIVE PROPERTIES CO522
Source: Patsnap Eureka. Assignee ranking and totals. Derived from a Patsnap search on Die-to-Die Interface Energy Efficiency covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP

Let an AI agent run this analysis on your own technology

Pick a task. Every answer cites the patents behind it.

10,000 free credits to start
The Numbers

Filing trend and technology composition

Two views of the same 114,847-record dataset: how filing activity has moved year over year, and which IPC subclasses carry the claim density.

A filing cycle that peaked in 2022

Annual filings rose from 700 in 2017 to a peak of 1,022 in 2022, then declined toward 46 in the most recent, still-partial year. The midpoint sits at the 2022 peak, meaning the second half of the window shows flat-to-declining activity rather than continued growth — a signal that the core interconnect approaches have largely been staked out.

A filing cycle that peaked in 202203006009001,200700201720182019202020211,0222022202320242025462026Most recent year is partial — publication lag means later filings are not yet visible.

Semiconductor devices dominate, but the tail is wide

H01L (semiconductor devices) leads at 5.6% of the 114,847 records in scope, with H10W following at 2.5%. Because a single record can carry several IPC classes, these shares add up to more than 100% and should be read as relative claim density across subclasses, not as a partition of the dataset. Notably, packaging and mechanical classes — B29C, B21D, B22D — each carry more filing weight than G11C, pointing to how much of this field's patent activity sits in physical assembly rather than pure logic design.

Semiconductor devices dominate, but the tail is wideH01L · Semiconductor devices6,4585.6%H10W2,9062.5%B29C · Shaping of plastics2,5022.2%B21D · Sheet-metal working1,7211.5%G06F · Electric digital data processi…1,2831.1%H10P1,1261.0%B22D · Metal casting1,1171.0%G11C · Static & digital memories6420.6%Other16,40914.3%

Shares are the percentage of the 114,847 records in scope. A patent can carry several IPC classes, so the shares add up to more than 100%.

Source: Patsnap Eureka. Filing trend and technology composition. Derived from a Patsnap search on Die-to-Die Interface Energy Efficiency covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.

Go deeper on Die-to-Die Interface Energy Efficiency with Eureka

This page is one run against one query. Ask Eureka your own question about die-to-die interface energy efficiency and every answer comes back with the patent numbers behind it.

Try Eureka
Representative Filing

A recent claim worth reading in full

Filed 2026-01-01
US20260003749A12026-01-01

Configurable die-to-die lane repair in multi-die systems coupled using link macros

MICROSOFT TECHNOLOGY LICENSING, LLC

Systems and methods for configurable die-to-die lane repair in multi-die systems are described. A multi-die system includes a first die and a second die, each comprising modular D2D link macros with M data lanes. The method forms repair groups spanning D data lanes, fewer than or equal to M, associated with one or more link macros, with D independently configurable per repair group, and designates R redundant lanes per repair group, R being independently configurable.Filed by Microsoft Technology Licensing, LLC — abstract trimmed for length.

US20260003749A1 — patent drawing 1US20260003749A1 — patent drawing 2
View full filing
Most-cited records in this dataset
#Publication no.Patent titleCitations
1US6271469B1Direct build-up layer on an encapsulated die package918
2US20060022214A1LED package methods and systems853
3US20110219208A1Multi-petascale highly efficient parallel supercomputer826
4US20110027458A1Continuous analyte sensors and methods of making same647
5US20110024043A1Continuous analyte sensors and methods of making same633
6US20110027453A1Continuous analyte sensors and methods of making same629
7US6405733B1Device for accurately marking tissue609
8US6155699AEfficient phosphor-conversion led structure577
9US6908784B1Method for fabricating encapsulated semiconductor components427
10US5374245AReinforced multiple-lumen catheter and apparatus and method for making the same425

Citation counts favour older filings simply because they have had more time to accumulate citations within this corpus — treat them as a signal of influence on the field, not of current technical importance.

Each row carries its publication number; clicking a row searches Eureka by that number.

Source: Patsnap Eureka. Citation counts and representative records. Derived from a Patsnap search on Die-to-Die Interface Energy Efficiency covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
Run it yourself

Put your own technology through the same analysis

 
Where to run it
Fastest

Eureka on the web

When you want the answer in the next five minutes.

The agent works the prompt against patents and technical literature, citing every source.

Run your analysis now →
For builders

MCP server & REST API

When it has to run inside your own pipeline.

Patent search, landscape analysis and assignee resolution as MCP tools. Drop them into any agent framework, or call REST directly.

Browse MCP servers →
Insights

Reading the concentration and momentum data together

Three figures matter more than the headline record count: how thin the leadership concentration actually is, how fast the largest filers are pulling back, and where the packaging-heavy IPC mix points next.

Concentration
8.4%
of all 114,847 records held by the top five

No single assignee controls the space

The top five ranked assignees combined account for 8.4% of all 114,847 records in scope, and the top ten only reach 11.2%. For a field with this much filing volume, that is a thin concentration — most of the claim space sits with entities outside the ranked leaders.

Based on the 100-company assignee ranking returned by the dataset.
Momentum
-79% to -97%
YoY change among the most active named filers

The largest filers are pulling back sharply

Every one of the most active assignees tracked for recent-year momentum shows a steep year-on-year decline, with drops as large as 97%. Combined with the 2022 peak in the overall trend, this suggests the core technical approaches in this generation of die-to-die interfaces have already been claimed by their originators.

Recent-year momentum reflects a partial final year; treat the direction, not the exact magnitude, as the signal.
Composition
5.6% / 2.5%
H01L and H10W shares of 114,847 records

Packaging classes carry real weight

H01L leads IPC composition, but B29C, B21D and B22D — plastics shaping, sheet-metal working and metal casting — each out-rank G11C memory classifications. That points to how much of the patented value in this space sits in the physical packaging and assembly of dies, not only in logic or protocol design.

IPC shares are measured against the 114,847-record total and sum to more than 100% because records carry multiple classes.
Eureka AI Agent
Looking for what nobody has claimed yet?

Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to die-to-die interface energy efficiency, with the prior art for and against each one.

Find the white space →
Source: Patsnap Eureka. Co-assignee relationships and derived observations. Derived from a Patsnap search on Die-to-Die Interface Energy Efficiency covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
Players

Who is filing, and where the field is still open

The ranked leaders are large semiconductor and systems companies, but their combined share of the dataset is small, and momentum data shows even the most active among them slowing sharply in the latest year.

Leader
3,740
records held by the top-ranked assignee

A clear leader, but not a dominant one

The leading assignee holds 3,740 records against a fifth-place figure of 852 and a tenth-place figure of 522 — a steep drop-off after the leader, then a long, gradually declining tail across the remaining ranked companies.

Ranking covers 100 companies, counted in records.
Co-filing
10 pairs
co-assignee relationships identified

Collaboration is limited and concentrated

Only ten co-assignee pairs appear in the dataset, and the strongest of them links two related entities under a shared corporate umbrella. Cross-company collaboration on filings is the exception here, not the norm.

Strongest pairing recorded 18 shared filings.
Receiving offices
11,820
United States filings, the largest single office

Filing activity is US-centric with a European and PCT tail

The United States receives more than three times the filings of the next-largest office, Europe (EPO) at 3,118, with WIPO's PCT route, the United Kingdom, Canada and Australia trailing behind. Strategy built around US prosecution alone would miss a meaningful share of the activity captured through PCT and EPO routes.

Figures count receiving-office filings, not granted patents.
🔍
Under-claimed sub-areas worth checking before filing
These pockets show comparatively thin claim density relative to the core H01L and H10W classes.
Protocol-layer overhead reduction schemesCross-vendor lane repair standardsStatic memory (G11C) interconnect couplingMetal-casting-based package reach solutionsStandard package reach qualification methods
Rank all filers by momentum →
Recent-year filing momentum among the largest assignees
AssigneeRecent yearYoY
Intel Corporation9-79%
Micron Technology, Inc.5-91%
Taiwan Semiconductor Manufacturing Co., Ltd. (TSMC)4-91%
Texas Instruments Incorporated1-86%
Samsung Electronics Co., Ltd.1-97%
Hewlett Packard Development Company, LP0
LG Energy Solution, Ltd.0-100%
Qualcomm Incorporated0-100%
Source: Patsnap Eureka. Assignee-level momentum. Derived from a Patsnap search on Die-to-Die Interface Energy Efficiency covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
What's Next

Where to take this next

The dataset points to a field with declining top-line filing activity but thin leadership concentration — which changes how a freedom-to-operate or whitespace exercise should be scoped.

Check packaging classes, not just logic

Given how much filing weight sits in B29C, B21D and B22D, a search limited to G06F and G11C will miss a large share of relevant prior art in physical die assembly and packaging.

Explore packaging-class filings in Eureka

Read the momentum data before assuming a leader is still active

The steep year-on-year declines among the most active assignees suggest some of the largest historical filers may be de-prioritising this exact claim space — worth confirming before treating them as blocking competitors.

Track assignee momentum in Eureka
Source: Patsnap Eureka. Forward-looking reading of the same dataset. Derived from a Patsnap search on Die-to-Die Interface Energy Efficiency covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
FAQ

Common questions about die-to-die interface energy efficiency patents

Answers are grounded in the same dataset. Derived from a Patsnap search on Die-to-Die Interface Energy Efficiency covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP

Research Die-to-Die Interface Energy Efficiency in depth with Eureka

Go past this page: query the whole die-to-die interface energy efficiency corpus yourself, in your own scope.
Every answer comes back with patent numbers you can open.

Try Eureka

Disclaimer. This page is generated from Patsnap Eureka data drawn from a limited snapshot of global patent and scientific-literature records, and is provided for general information and reference only.

Patent data carries inherent limitations: recent filings (typically the most recent 18–24 months) are under-counted due to standard publication lag; counts may be reported at either a patent-family or a patent-record basis and are not always directly comparable; classification, applicant-name, and citation data may contain errors, duplicates, or omissions; and the underlying search query defines and constrains the scope shown. As a result, the analysis may be incomplete or inaccurate and may not reflect the full technology landscape.

Nothing on this page constitutes an exhaustive prior-art, novelty, freedom-to-operate, or validity search, nor does it constitute legal, financial, investment, or professional advice, and it should not be relied upon as such. Any patent, commercial, or strategic decision should be verified independently and reviewed with qualified patent, legal, and domain professionals. Patsnap makes no warranties, express or implied, as to the accuracy, completeness, or fitness for any particular purpose of the information presented.

Machine translation. Assignee and organisation names originally recorded in Chinese, Japanese or Korean have been rendered into English by an AI translation step so that the tables stay readable. These renderings are best-effort and may not match a company’s registered English name; the original name is what the underlying patent record carries, and it is what any Eureka query launched from this page uses.

Help us improve this page

Found incorrect or outdated information? Let us know and we'll get it fixed.