Invest1 distinct publisher2 min readPublished
Weekly x402 stablecoin transfers doubled and Base's share of them fell to 48 percent. The dollars behind those transfers annualize to about $19 million.
The Investor · Invest desk
Compiled by The InvestorSomething wrong?How this is made
Divide each chain's dollars by its transfer count and the category splits into two unrelated businesses. Base moved $191,816 across 4.2 million transfers, about 4.6 cents each [2]. Solana's 3.3 million transfers carried $91,715, roughly 2.8 cents [3]. Polygon worked out near 3.1 cents [4]. Algorand looks nothing like the others: 109,000 transfers carrying $50,785, about 47 cents apiece [5]. It was 1.3 percent of the week's transfer count and 13.8 percent of its value [6].
That spread is what the diversification actually consists of. Base holds 52 percent of the dollars against 48 percent of the count, while Solana carries 38 percent of the count for 25 percent of the dollars [11]. The traffic that has left Base is the cheapest traffic in the set. Run the same division on the November 2025 peak and the average transfer then was about 50 cents [7], roughly twelve times what a transfer carried in the week of August 17 [8]. The count has recovered to 43 percent of that peak; the money has recovered to 3.7 percent [9].
At $367,950 a week, the whole measured rail annualizes to about $19 million if the pace holds [10]. Ten basis points on every transfer on every chain would be around $19,000 of annual revenue [13]. That is the number anyone selling into this market has to plan around, and it says the flow is not the product. Gateways, metering and API access are.
The protocol's design explains the shape. Machines pay per request, with no card rail, invoice or subscription tier sitting in between [11], so each transfer is a billing event attached to a call. The count behaves like a usage metric, not gross merchandise value, which is why it can double in a week [1] without the dollar line following.
If Algorand's 47 cent average holds as its counts grow, it is the only place in this data where agents are paying for something priced above an endpoint ping. Everything else is sub-nickel. And the base rate deserves respect: the same series sat near 2 million for most of the year and dropped below 1 million in March before the June turn [3], with every transfer in it denominated in USDC [12]. A number that doubles in seven days halves in seven too.
Ranked by verification strength, evidence, and original report placement.
Token Terminal data showed 8.7 million x402 stablecoin transfers in the week starting August 17, more than double the 4.1 million recorded the week prior, making it the busiest week year to date.
The week before the 8.7 million reading recorded 4.1 million x402 stablecoin transfers.
For most of the year weekly x402 transfer counts were near 2 million and bottomed below 1 million in March; the uptick started in June and has held since.
The category's all-time high was around 20.1 million transfers in the week of November 17, 2025.
In the week of November 17, 2025, USDC on Base accounted for 18.7 million of the 20.1 million transfers, roughly 93 percent, with Solana at 1.3 million and Polygon at 127,800.
In the week starting August 17, Base handled 4.2 million transfers for a 48 percent share, Solana 3.3 million for 38 percent, Polygon 1.1 million and Algorand 109,000.
Distinct publishers with included, body-backed reporting in this cluster.
1 article · August 25, 2026
Follow any of these and your For You feed starts watching them — no settings page required.
build
France's e-invoicing deadline is really a bill for supplier data nobody verified1 distinct publisher
invest
OpenAI and AWS just made agent-initiated payments a documented pattern, not a demo1 distinct publisher
invest
USDC clears 77% of on-chain transfer volume, and its record DEX day is a rounding error1 distinct publisher
invest
X's stablecoin creator payouts would be a plumbing decision, not a crypto bet3 distinct publishers
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Quantitative but single-sourced
Every figure is specific and internally consistent (counts, dollar volumes and chain splits reconcile with the stated averages), but all of it flows from one analytics provider, Token Terminal, reported by one outlet with no link to the underlying dashboard, no methodology note on what x402 traffic is counted, and no second reading of the same series. The interpretive claim about Solana's rise reflecting real deployment has no supporting payer-level data.
Real usage, negligible economic weight
Adoption of the mechanism is genuine and measurable: 8.7 million transfers in a week, a sustained climb since June, and traffic now spread across four chains rather than concentrated on Base. Economic adoption is minimal, however — $367,950 of weekly value, about 4 cents per transfer, roughly $19 million annualized, and only about 3.7 percent of the November 2025 dollar peak. No named production users, counterparties or payer counts are disclosed.
Milestone framing runs slightly ahead of the dollars
The headline framing ('AI Agent Payments Hit 2026 High') and the doubling of transfer counts read as a growth breakout, while the underlying flow is under $400,000 a week at 4 cents per transfer. The gap is small rather than large because the source itself discloses and argues the shortfall — it states the volume is not even a year-to-date high and concludes the chart shows 'adoption of a mechanism, not the arrival of an economy.'
Crypto-native outlet, ecosystem-favourable subject
The only source is a crypto-native publication and the author is described as its own research analyst and host of its market streams, so milestone-shaped coverage of onchain activity aligns with the outlet's audience interest. The story's named beneficiaries — Circle's USDC monopoly, Base, and a fast-gaining Solana — are ecosystems whose share figures the piece amplifies. No positions, sponsorships, or data-provider relationships are disclosed either way, and the piece's cautionary conclusion works against a purely promotional read.
Moderate-low
Figures are precise, mutually consistent and drawn from a recognised onchain analytics provider, and the derived ratios follow directly from them, which supports the descriptive core. Confidence is held down by total reliance on one publisher and one dataset, absent methodology, and no payer-level data to test whether the transfer surge represents broad agent deployment or concentrated machine traffic.