Matter Certification Statistics

Every Matter device has to be certified by the Connectivity Standards Alliance, and each certification is published on a public ledger, the Distributed Compliance Ledger (DCL). This page counts what is in that ledger: how many end-product models are certified, when they first appear, which vendor IDs certify the most, which network transports they declare, and which Matter specification version their latest certification records.

It is a global analysis of the certification ledger, computed by Products That Matter from a complete snapshot of it. It is not a count of products sold, shipped or available in stores. The caveats follow the key figures; data coverage and methodology are at the end.

Latest qualifying certification date
Nov 10, 2026
Data captured
Nov 11, 2026, 10:52 CET (UTC+01:00)
Statistics updated
Nov 11, 2026, 10:52 CET (UTC+01:00)
Ledger block height
330,275

Times are Europe/Budapest. Certification dates are calendar dates as the ledger records them. A recent capture does not mean recent certification dates.

This snapshot:JSONCSV

On this page

Key figures

From the ledger snapshot captured Nov 11, 2026

Main finding · year to date

Jan 1–Nov 10, 2026 vs. Jan 1–Nov 10, 2025

Newly certified models are up 65.8% on the same window of 2025, and up 21.9% without the top 5 vendor IDs.

All vendor IDs

+65.8%

2025 3,017
2026 5,001

Excluding the top 5 vendor IDs

+21.9%

2025 2,062
2026 2,513

The top 5 vendor IDs on their own

+160.5%

2025 955
2026 2,488

Newly certified models in each window. Both windows end on the same calendar day, and all six bars share one scale. The top 5 are the vendor IDs with the most certified models in the whole snapshot: Quectel, Signify, Aqara, AiDot and Top Victory Investments Limited. Their models are removed from both windows. Vendor IDs are not brands or corporate groups.

Certified end-product models

13,142

Distinct vendor ID / product ID pairs

Per year →

Vendor IDs represented

497

With at least one certified model

Vendor IDs →

Q3 2026 vs. Q3 2025

−26.7%

966 vs. 1,318 newly certified models; both completed quarters

Per quarter →

Latest certification below Matter 1.4

45.5%

5,982 of 13,142 models; 2 with no known version

Specification versions →

Declared network transports

Transport details →

Share of the 13,011 models with a known transport. A model can declare more than one, so the shares overlap and add up to more than 100%.

Wi-Fi75.5%
9,820 models
Thread21.4%
2,784 models
Ethernet5.5%
714 models

Read before quoting

What these numbers do not tell you

  • A model ID is not a unique retail design. Variants, regional SKUs, and OEM or rebranded products can have their own IDs.
  • Vendor IDs are not independent corporate groups and are not necessarily consumer brands. Module makers and OEMs certify on behalf of others.
  • Certifications are not sales, shipments, adoption, installed firmware or product launches. They are not proof that a product is available in the US or on Amazon.
  • The snapshot holds the qualifying records that exist today. Backfills, status changes and source corrections can revise past cohorts, so earlier years can change between updates.
  • Declared transports are a union over all certified software versions, so they can include declarations from older software.
  • The latest certified Matter version is not the firmware installed on a device, a promise of upgrades or a guarantee of platform features.
  • The Products That Matter shopping catalog is a curated selection. It is not a representative census of what is available, and it is not the source of these numbers.

Newly certified models per year

Each model counts once, in the year of its earliest retained certification date. Matter certification records in this snapshot start on Oct 24, 2022.

2022
Full year · records start Oct 24, 2022
418
vs. 2021 full year: 0
N/A
2023
Full year
2,061
vs. 2022 full year: 418
+393.1%
2024
Full year
1,749
vs. 2023 full year: 2,061
−15.1%
2025
Full year
3,913
vs. 2024 full year: 1,749
+123.7%
2026 to date
Partial: Jan 1–Nov 10, 2026
5,001
vs. Jan 1–Nov 10, 2025: 3,017
+65.8%

Figure 1. Change is (this period / comparison period − 1) × 100. N/A means the comparison period has no models. The tick on 2025 marks its Jan 1–Nov 10 count, the comparison for 2026.

Newly certified models per quarter

Each quarter is compared with the same quarter a year earlier. Q4 2026 is still in progress, so it is compared with the same days of Q4 2025 only.

Figure 2. Newly certified models per quarter, Q4 2022–Q4 2026. Q4 2026 covers Oct 1–Nov 10, 2026 only and is hatched; it is not comparable with a completed quarter. The table has the same numbers.
Full table: 17 quarters, each with the same quarter a year earlier
Newly certified models per quarter, share of the year and change from the same quarter a year earlier
Quarter Window Models Share of year Same quarter a year earlier Change
Q4 2026 Partial: Oct 1–Nov 10, 2026 625 12.5% of YTD Oct 1–Nov 10, 2025: 740 −15.5%
Q3 2026 Jul 1–Sep 30, 2026 966 19.3% of YTD Jul 1–Sep 30, 2025: 1,318 −26.7%
Q2 2026 Apr 1–Jun 30, 2026 1,875 37.5% of YTD Apr 1–Jun 30, 2025: 628 +198.6%
Q1 2026 Jan 1–Mar 31, 2026 1,535 30.7% of YTD Jan 1–Mar 31, 2025: 331 +363.7%
Q4 2025 Oct 1–Dec 31, 2025 1,636 41.8% Oct 1–Dec 31, 2024: 606 +170.0%
Q3 2025 Jul 1–Sep 30, 2025 1,318 33.7% Jul 1–Sep 30, 2024: 309 +326.5%
Q2 2025 Apr 1–Jun 30, 2025 628 16.0% Apr 1–Jun 30, 2024: 445 +41.1%
Q1 2025 Jan 1–Mar 31, 2025 331 8.5% Jan 1–Mar 31, 2024: 389 −14.9%
Q4 2024 Oct 1–Dec 31, 2024 606 34.6% Oct 1–Dec 31, 2023: 910 −33.4%
Q3 2024 Jul 1–Sep 30, 2024 309 17.7% Jul 1–Sep 30, 2023: 393 −21.4%
Q2 2024 Apr 1–Jun 30, 2024 445 25.4% Apr 1–Jun 30, 2023: 501 −11.2%
Q1 2024 Jan 1–Mar 31, 2024 389 22.2% Jan 1–Mar 31, 2023: 257 +51.4%
Q4 2023 Oct 1–Dec 31, 2023 910 44.2% Oct 1–Dec 31, 2022: 418 +117.7%
Q3 2023 Jul 1–Sep 30, 2023 393 19.1% Jul 1–Sep 30, 2022: 0 N/A
Q2 2023 Apr 1–Jun 30, 2023 501 24.3% Apr 1–Jun 30, 2022: 0 N/A
Q1 2023 Jan 1–Mar 31, 2023 257 12.5% Jan 1–Mar 31, 2022: 0 N/A
Q4 2022 Oct 1–Dec 31, 2022 418 100.0% Oct 1–Dec 31, 2021: 0 N/A

Share of year for the current year is of year to date. N/A means the quarter a year earlier has no models.

Latest quarters

  • Q4 2026 625 Partial: Oct 1–Nov 10, 2026 12.5% of YTD Year earlier: 740Oct 1–Nov 10, 2025 −15.5%
  • Q3 2026 966 Jul 1–Sep 30, 2026 19.3% of YTD Year earlier: 1,318Jul 1–Sep 30, 2025 −26.7%
  • Q2 2026 1,875 Apr 1–Jun 30, 2026 37.5% of YTD Year earlier: 628Apr 1–Jun 30, 2025 +198.6%
  • Q1 2026 1,535 Jan 1–Mar 31, 2026 30.7% of YTD Year earlier: 331Jan 1–Mar 31, 2025 +363.7%
Show 13 earlier quarters (Q4 2022–Q4 2025)
  • Q4 2025 1,636 Oct 1–Dec 31, 2025 41.8% Year earlier: 606Oct 1–Dec 31, 2024 +170.0%
  • Q3 2025 1,318 Jul 1–Sep 30, 2025 33.7% Year earlier: 309Jul 1–Sep 30, 2024 +326.5%
  • Q2 2025 628 Apr 1–Jun 30, 2025 16.0% Year earlier: 445Apr 1–Jun 30, 2024 +41.1%
  • Q1 2025 331 Jan 1–Mar 31, 2025 8.5% Year earlier: 389Jan 1–Mar 31, 2024 −14.9%
  • Q4 2024 606 Oct 1–Dec 31, 2024 34.6% Year earlier: 910Oct 1–Dec 31, 2023 −33.4%
  • Q3 2024 309 Jul 1–Sep 30, 2024 17.7% Year earlier: 393Jul 1–Sep 30, 2023 −21.4%
  • Q2 2024 445 Apr 1–Jun 30, 2024 25.4% Year earlier: 501Apr 1–Jun 30, 2023 −11.2%
  • Q1 2024 389 Jan 1–Mar 31, 2024 22.2% Year earlier: 257Jan 1–Mar 31, 2023 +51.4%
  • Q4 2023 910 Oct 1–Dec 31, 2023 44.2% Year earlier: 418Oct 1–Dec 31, 2022 +117.7%
  • Q3 2023 393 Jul 1–Sep 30, 2023 19.1% Year earlier: 0Jul 1–Sep 30, 2022 N/A
  • Q2 2023 501 Apr 1–Jun 30, 2023 24.3% Year earlier: 0Apr 1–Jun 30, 2022 N/A
  • Q1 2023 257 Jan 1–Mar 31, 2023 12.5% Year earlier: 0Jan 1–Mar 31, 2022 N/A
  • Q4 2022 418 Oct 1–Dec 31, 2022 100.0% Year earlier: 0Oct 1–Dec 31, 2021 N/A

Share of year for the current year is of year to date. N/A means the quarter a year earlier has no models.

How each year splits by quarter

Each quarter's newly certified models as a percentage of that year's total. For 2026 the total is year to date (through Nov 10, 2026), so its shares cover only the quarters so far and are not comparable one-for-one with full years.

Figure 3. Each quarter's share of its year. Current-year shares are of the year to date through Nov 10, 2026, so they are not comparable one-for-one with full years. Quarter labels are printed in or under every bar.

Year × quarter table
Newly certified models per quarter as a share of each year's total
Year Q1Q2Q3Q4 Total
2022 – – – 100.0%418 418
2023 12.5%257 24.3%501 19.1%393 44.2%910 2,061
2024 22.2%389 25.4%445 17.7%309 34.6%606 1,749
2025 8.5%331 16.0%628 33.7%1,318 41.8%1,636 3,913
2026 YTD 30.7%1,535 37.5%1,875 19.3%966 12.5%625 (partial) 5,001

A dash means the quarter is before the first certification record or has not started yet. Q4 2026 is partial: Oct 1–Nov 10, 2026.

Vendor IDs with the most certified models

Ranked by certified models across the whole snapshot. Names come from the ledger's vendor directory, matched by exact numeric vendor ID. This counts certifications only. Vendor IDs are not brands or corporate groups, and one company can certify for other brands.

Top 5 vendor IDs 4,893 models, 37.2% of all certified models. Removed in the “excluding the top 5” comparison.
1 Quectel ID 5136 (0x1410) 1,879 14.3%
2 Signify ID 4107 (0x100B) 1,227 9.3%
3 Aqara ID 4447 (0x115F) 641 4.9%
4 AiDot ID 5014 (0x1396) 639 4.9%
5 Top Victory Investments Limited ID 4952 (0x1358) 507 3.9%
Ranks 6–10
6 tcl-vendor ID 4916 (0x1334) 505 3.8%
7 Tuya ID 4701 (0x125D) 396 3.0%
8 Uascent ID 5120 (0x1400) 386 2.9%
9 Longan.link ID 4895 (0x131F) 353 2.7%
10 CoolKit ID 4742 (0x1286) 337 2.6%

Figure 4. 6 vendor IDs (7 models) have no name in the directory. They stay in every total. Ties are ordered by lower vendor ID. The full ranking is in the downloads.

Wi-Fi, Thread and Ethernet

The network transports each model declares across all its certified software versions. Bluetooth is left out because Matter uses it only for setup. The 131 models that declare none of these are unknown, not unconnected.

Wi-Fi 9,820 models 75.5%
Thread 2,784 models 21.4%
Ethernet 714 models 5.5%
Overlap already counted in both rows above
Both Wi-Fi and Thread 122 models 0.9%
Known transport (denominator)
13,011
Unknown transport (declares none)
131
All certified models
13,142

Figure 5. Each bar is out of 100% of the 13,011 models with a known transport. A model counts once for each transport it declares, so the shares overlap and add up to more than 100%.

Matter specification versions

The specification version recorded on each model's latest certification date. 5,982 of 13,142 models (45.5%) record a known version below 1.4. That is 45.5% of the 13,140 with a known, unambiguous version. A recorded version is certification metadata. It does not tell you the firmware installed on a device, promise an upgrade or guarantee platform features.

Below Matter 1.4 5,982 models, 45.5%
1.0.0 1,558 11.9%
1.1.0 780 5.9%
1.2.0 943 7.2%
1.3.0 2,701 20.6%
Matter 1.4 and later 7,158 models, 54.5%
1.4.0 5,283 40.2%
1.4.1 468 3.6%
1.4.2 368 2.8%
1.5.0 882 6.7%
1.5.1 155 1.2%
1.6.0Rare value 2 <0.1%
No version assigned
Unknown (zero or missing) 2 <0.1%
Ambiguous (records on the same latest date disagree) 0 0.0%
Anomalous value (not a valid version encoding) 0 0.0%

Figure 6. Gray bars are versions below 1.4; the group headings carry the same split in text. “Rare value” marks versions recorded for fewer than 10 models. Treat those with caution. A version showing up in the ledger is not an announcement that it has launched.

Reference

Data coverage

Records in the ledger snapshot and how many were used
Compliance records in the snapshot 15,733
Retained (Matter, certified, end product) 15,699
Excluded: software components 27
Excluded: program type missing or other 7
Excluded: provisional, revoked or other status 0
Excluded: not a Matter certification 0
Retained records with an unusable date (missing, invalid, future or before 2020) 0
Certified models without any usable date (in totals, not in yearly counts) 0
Certified models without a public model-detail record 7,916 of 13,142
Vendor directory entries in the snapshot 898

The model-detail records (names, device types) cover only some certified models, so this page ranks no device categories. Every model is counted from its compliance records, whether or not it has a detail record. The vendor directory lists 898 entries, but only 497 vendor IDs have a certified end-product model.

Methodology

Methodology version 1.0.0
  1. One complete snapshot. Every compliance record, model-detail record and vendor-directory entry is read from the ledger’s public API at one block height. A capture is used only if every page came from that height and the row counts match the totals the ledger reports. Otherwise the previous snapshot stays live.
  2. Which records count. Only records with certification type “matter”, status certified and program type “endProduct”. Software components, records without a program type, and provisional or revoked records are left out.
  3. What a model is. One distinct numeric vendor ID and product ID pair, however many software versions or certificates it has.
  4. Cohort date. A model’s earliest retained certification date: the first observed currently certified end-product record. Dates are used as the calendar dates the ledger records, with no time-zone shift. Missing, impossible, future-dated and pre-2020 dates are reported in the data coverage and are not used.
  5. Year-over-year comparisons. Change = (this period / comparison period − 1) × 100. With no models in the comparison period it is N/A. A partial year or quarter is compared with the same calendar window a year earlier, ending on the same month and day as the latest qualifying certification date. February 29 maps to February 28 in a non-leap comparison year. Full years and completed quarters are compared with the full prior period.
  6. Quarterly share of a year. A quarter’s newly certified models divided by its year’s total. For the current year the total is year to date.
  7. Growth excluding the top five vendor IDs. The same matched-window comparison, after removing every model from the five vendor IDs with the most certified models in the whole snapshot (ties go to the lower vendor ID). The same five are removed from both windows.
  8. Vendors. Distinct vendor IDs of the counted models, named by exact vendor ID from the ledger’s directory. IDs without a name stay in all totals. No corporate groups are merged.
  9. Transports. The union of Wi-Fi, Thread and Ethernet over all of a model’s retained records. Spelling is normalized (“wi-fi” and “wifi” count the same) and Bluetooth is ignored. The shares are of models declaring at least one of the three and are not exclusive.
  10. Specification version. The version on the model’s latest retained certification date, decoded from the ledger’s 32-bit value (major, minor, patch, reserved). Zero or missing values are unknown. If records on that date disagree, the model is ambiguous rather than assigned a version. Values with a nonzero reserved byte or a zero major version are anomalous.

Sources and downloads

Source

Connectivity Standards Alliance Distributed Compliance Ledger: web interface, plus the public REST endpoints compliance-info, models and vendors.

The ledger data is published by the Connectivity Standards Alliance, and its terms apply to the source data. Products That Matter does not grant or claim a license for it.

Downloads

Aggregate data from the same snapshot as this page, with source and methodology details.

Download JSON Download CSV

Spotted a problem? Tell us.