Connect your own processor

Two shapes of rail can sit on one TensorRail account, and they behave differently in exactly one respect: where the money lands. Everything else — one integration, one payment object, one set of webhooks, one dashboard — is identical.

The two shapes

TensorRail solutions

TensorRail enables the rail. You do not integrate with, contract with, or hold an account at any underlying institution; TensorRail presents these as its own rails. A collection on one of these rails is recorded to your TensorRail balance on the ledger.

This is the default, and for most accounts it is the only shape in use.

Your own processors

You already hold an account with a payment processor and you want to keep using it. You enter that processor's credentials in your dashboard, and TensorRail routes payments through it on your behalf.

The funds land in your processor account, not with TensorRail — so no TensorRail balance is credited for those payments, and they do not appear in your TensorRail settlement figures. You keep the commercial relationship, the pricing, and the payout schedule you already have with that processor. What you gain is one integration and one dashboard across both shapes.

Connecting one

Settings → Processors → Add a processor. Pick the processor, enter its credentials, save.

Available today:

Processor	Region	Methods
Stripe	Global	Cards, Apple Pay, Google Pay, Link, SEPA
Adyen	Global	Cards, Apple Pay, Google Pay, iDEAL, Klarna, SEPA, Bancontact
Checkout.com	Global	Cards, Apple Pay, Google Pay
Mollie	Europe	Cards, iDEAL, Bancontact, SOFORT, Klarna, EPS
Razorpay	India	Cards, UPI, Netbanking, Wallets
Rapyd	Global	Cards, Bank transfer, Wallets
CryptoPay	Crypto	Bitcoin, Ethereum, Litecoin
BitPay	Crypto	Bitcoin, Ethereum, USDC, Litecoin

Where the credentials come from is your processor's own dashboard. TensorRail never sees more than what you paste in, and the secret half is never shown back to you after saving.

The inbound webhook leg — do not skip this

This is the step most easily missed, and skipping it does not fail loudly. It leaves payments sitting unresolved.

TensorRail tells your processor to send payment updates to a URL that is unique to your account and that connector. You must register that URL in your processor's own dashboard. It is shown on the processor's card in Settings → Processors once the connector is saved.

Six of the eight processors above — Stripe, Adyen, Checkout.com, Rapyd, CryptoPay and Razorpay — also verify the webhooks they send, using a signing secret you configure at the processor. If that secret is not configured, or does not match, TensorRail rejects the delivery rather than trust it. That rejection is deliberate: an unverified payment notification is the one thing that must never be believed.

The callback URL is configured at your processor, not here

Saving credentials in TensorRail does not register the callback. Until you set it in your processor's dashboard, payments will start but never reach a final status in TensorRail, because nothing is telling us they completed.

What stays the same

• One API. The same POST /payments, the same payment object, the same next_action.
• One set of webhooks. Your endpoint receives the same events regardless of which shape handled the payment.
• One dashboard. Both shapes appear in the same transaction list.
• Test first. A connector can be saved in test mode and exercised end to end before it takes real money.

What differs

	TensorRail solutions	Your own processor
Who holds the processor account	TensorRail	You
Where the funds land	Your TensorRail balance	Your processor account
Credited to your TensorRail balance	Yes	No
Who sets the processor pricing	TensorRail	You, with your processor
Inbound callback URL	Configured for you	You register it at your processor

If you are unsure which shape a rail on your account uses, the Processors screen separates them under two headings: TensorRail solutions and Your own processors.