Measured directly

Market overview

Base blocks are read in full; Solana is read from known facilitators. Every row maps to a specific transaction — nothing is extrapolated or filled in. Window: 30 days.

Payments
249,501
2,153 in 24h
Median ticket
$0.00549
mean $7.14
Payers
9,664
60% come back
Sellers with flow
338
9,818 recipients total

Why the mean ticket is useless here

95%
of volume comes from payments above $100

There are 1,022 such payments — 0.4% of the count. These are ordinary gasless stablecoin transfers, not API calls: a few hundred dollars has nothing to do with machine commerce.

So our headline figure is the median, $0.00549. The mean ($7.14) sits beside it for completeness. Any market estimate built on the mean ticket is off by orders of magnitude.

Ticket size distribution

TicketPaymentsShare of countShare of volume
≤ $0.01 159,763 64% <0.1%
$0.01–0.10 73,382 29% 0.1%
$0.10–1 8,054 3% 0.3%
$1–100 7,280 3% 5%
> $100 1,022 0.4% 95%

The two right-hand columns tell the whole story: nearly every payment is tiny, while nearly all the volume sits in a handful of large transfers.

Who pays

Payers
9,664
unique addresses
Returning
5,772
three payments or more
Agent-like
1,503
regular intervals, ≥70%
Circular
574
both pay and receive
PayersAddressesShare
with ≥1 payments9,664 100%
with ≥3 payments5,772 60%
with ≥10 payments2,278 24%
with ≥100 payments277 3%

A one-off wallet is a test or a demo. Only someone who pays regularly can become a customer. Payers with scores →

Who gets paid

SellersCountShare
with ≥1 distinct payers 9,818 100%
with ≥3 distinct payers 338 3%
with ≥10 distinct payers 97 1.0%
with ≥50 distinct payers 26 0.3%
top 10 recipients, share of volume 91%
The “≥10” row is the real addressable base of sellers. Catalogues list tens of thousands of endpoints, but only a handful see actual flow: one operator often holds dozens of listings, so a catalogue count says nothing about how many businesses exist.
RecipientPayersPaymentsVolume
0x7284d41b…299c 1,30349,783 $995.66
0x68396bd3…1a2b 71513,430 $10,413.22
0x3e0a7c9a…78e7 6204,968 $21.32
0x4f4beb65…422d 564564 $6.1
0x8a12157b…0881 526526 $844.96
0xef4364fe…6c7f 4832,135 $21.35
0x159be9e3…4908 3201,534 $153.4
0x52e29e0d…f3ea 2942,186 $18.31
0x5aae638c…482e 238781 $195.25
0x9c955c40…e299 1972,482 $2.48

Who settles

Settling addresses
228
they pay the gas
Identified
13
listed in public registries
In no registry
215
94% of all
x402 is open — anyone can run a settlement service, and most never register anywhere. An explorer that walks catalogues misses these payments entirely. We find them by reading blocks. Full list →

Endpoint health

Probed
9,277
of 9,277 catalogued
Answer 402
84%
38,795 probes
Clean envelope
74%
strictly to the v2 schema
Median latency
169 ms

Grade distribution

Grade B
7,287
Grade C
434
Grade D
1,477

Endpoints without enough observations are left out — they get “not enough data”, not a low grade.

What breaks on the seller side

CodeObservations
E_NO_4025,535
E_NETWORK_NOT_CAIP22,072
E_EXTENSIONS_IN_ACCEPTS1,239
E_NO_RESOURCE467
E_REDIRECT_301329
E_PAYTO_DIFFERS_FROM_CATALOG285
E_ENVELOPE_IN_BODY266
E_VERSION_1216
E_UNREACHABLE213
E_PRICE_BELOW_QUOTE177
E_TESTNET_IN_PRODUCTION161
E_NO_ENVELOPE151
E_HTTP_PLAINTEXT132
E_REDIRECT_302123
E_PRICE_ABOVE_QUOTE108
E_NO_ACCEPTS86
E_PAYTO_CHANGED67
E_LEGACY_AMOUNT_FIELD26
E_REDIRECT_30725
E_REDIRECT_30813
E_REDIRECT_3034

Endpoints with findings →

By network

NetworkPaymentsMedianVolume PayersRecipients
Base239,534 $0.00500$1.8M 9,5549,717
Solana9,967 $0.01$371.97 110101

Polygon and Arbitrum are excluded on purpose: a full block census found four payments in two hours on Polygon (mean ticket $33, so not x402) and none at all on Arbitrum.

By day

DatePaymentsVolumePayers
2026-09-29 2,050$56.81 55
2026-09-28 1,117$105.28 60
2026-09-27 36,492$136K 3,015
2026-09-26 70,733$593K 4,353
2026-09-25 56,406$456K 4,405
2026-09-24 73,228$578K 4,825
2026-09-23 9,026$17,431.78 696
2026-09-22 449$337.92 100

Collectors

JobLast runResultNote
catalog9h ago ok scanned 21517, endpoints 9205 → 9277
census:base7h ago ok blocks 51857980..51858679 (700), transactions 94494, payments 1263, new 1263, lag 87816, unknown token 29, permit2 2
facilitators7h ago ok seed 66, live responses 9/16, addresses from /supported 85, total 409
probe7h ago ok probed 400, answered 402: 328, clean: 271
sanctions21h ago ok sources 3/3, addresses 1045, stored 1045 (EVM 126)
score7h ago ok endpoints 9198/9277, wallets 6557
solana7h ago failed

Same data for machines: /health · aggregates: /v1/market/stats

Next

Endpoints →

Scores, findings and the full observation history per resource.

Wallets →

Payers, recipients and facilitators, listed separately.

Facilitators →

Who actually settles — including those in no public registry.

API →

Verdict before payment, seller and wallet scores, market aggregates.

We hold no fundsNo custody, no escrow, no facilitator role. Not one payment passes through us.
Same input, same answerScoring is deterministic. No model, no learning on the fly.
Objections welcomeWe show owners what drags a score down. How to dispute
We say what we can't seeSome payments move off-chain, and we say so instead of pretending otherwise.