Agent leaderboards / All sectors / Payments

Payments: which providers coding agents choose

Stripe took about 88% of 657 runs across 11 codebases.

657 runs11 apps5 agents4 personasupdated 2026-09-29

The interactive board, open on payments. Open it full page · this page as Markdown

Read this leaderboard as textrankings, key learnings, method

Key learnings

We asked five coding agents to add payments to 11 codebases, 657 times in all. We varied the wording of the ask and who was doing the asking. Paddle came second with 27 wins.

26 of 55

Same codebase, same agent, different answers

We grouped runs by codebase and agent, then looked at whether the repeated asks agreed. In 26 of 55 groups they didn't all land on the same provider.

180 of 180

One persona never picked anything else

Junior developers chose Stripe in all 180 of their runs. Senior engineers gave Paddle 27 wins, and enterprise teams gave GoCardless 16.

9 of 244

Named in hundreds of runs, picked in nine

Adyen came up in 244 runs and was chosen in nine of them. PayPal came up in 206 runs and was never chosen. Square was named 150 times and won once.

The simulated user approved all 657 plans, but sent the agent back at least once in 83 runs.
In 67 runs it refused to approve until the agent named a specific provider.
The procurement and compliance framing didn't change the order, with Stripe taking 59 of those 60 runs.
All five agents put Stripe first, though their second picks differed: Mollie behind Cursor, Paddle behind Codex.
Bottomline PTX won three runs and AccessPay two, all in the .NET utility billing service.

Explore every run in the interactive board

The ranking657 runs

	Product	Wins	Share
1	Stripestripe.com	578	88%
2	Paddlepaddle.com	27	4%
3	Molliemollie.com	18	3%
4	GoCardlessgocardless.com	17	3%
5	Adyenadyen.com	9	1%
6	Bottomline PTXbottomline.com	3	0%
7	AccessPayaccesspay.com	2	0%
8	Helcimhelcim.com	2	0%
9	Squaresquareup.com	1	0%

By agent, by persona, by wording

By agent

Codex198 runs	Stripe · 169then Paddle · 18
Claude Code165 runs	Stripe · 155then Adyen · 4
Cursor164 runs	Stripe · 139then Mollie · 12
Muse Code · Muse Spark 1.397 runs	Stripe · 86then Paddle · 7
Grok Build CLI · Grok 4.733 runs	Stripe · 29then GoCardless · 2

By persona

Junior developer180 runs	Stripe · 180
Senior engineer180 runs	Stripe · 139then Paddle · 27
Enterprise team178 runs	Stripe · 148then GoCardless · 16
Vibe coder119 runs	Stripe · 111then Mollie · 5

By what the ask stressed

The plain ask597 runs	Stripe · 519then Paddle · 27
Procurement and compliance60 runs	Stripe · 59then GoCardless · 1

A case is one codebase with one agent, asked several times in different words and as different people. 26 of 55 cases did not hold to a single provider.

How this was measured

Every number on this page comes from a controlled experiment. We took 11 small applications, asked 5 coding agents (Codex, Claude Code, Cursor, Muse Code (Muse Spark 1.3), Grok Build CLI (Grok 4.7)) to add payments to each of them, in several wordings and as a junior developer and senior engineer and enterprise team and vibe coder, and let the agent choose the product. Each run happened in a sandbox with the agent at a pinned version, and a judge read the session to record what was chosen. That is 657 runs. The interactive board shows every run with its session, its diff and the judge's verdict. A simulated user stood in for the owner of the codebase: it read the agent's plan and had to approve it before any code was written; it sent the agent back at least once in 83 runs. Read the methodology and the publications.

If you sell in this sector: what these numbers mean for a vendor.

Open the interactive board This page as Markdown