llms.txt
Compare with other sites' llms.txt files →# seams
> Build the portal to launch your AI product. A gateway your customers call, model names you own, routing you control, metered and capped while the response is still streaming.
> Who may call, which model they reach, and what they can spend: one path. Metered and checked together on every request, including mid-stream.
> **Sign-up is open.** Get started at [console sign-in](https://console.ourseams.com/sign-in) with a GitHub account.
## What "portal" means here
Routing, billing, and identity usually live in different tools. Seams connects them so your customers get one portal and you get one place that enforces access, metering, and spend.
- **Access**: Decide which models each customer can reach, under names you choose. Change the target without redeploying clients.
- **Control**: Know who is calling and how much they can spend. Usage is metered on the request path; caps hold for the life of the stream.
- **Monetization**: Plans, balances, and your markup. Provider cost and what you charge on the same ledger.
You ship an AI product and need a gateway customers call, model names you own, metering, and spend limits that actually hold, without building that layer yourself. Whoever holds the key (your customer, their agent, or your team) uses one portal for usage and one gateway for what they may reach and spend.
## Sign-up
Sign in with GitHub at https://console.ourseams.com/sign-in. Onboarding creates an application, a provider connection, and a key. Configure aliases and bundles, mint end-user keys, and point clients at your gateway.
## Pages
- [Front page](https://ourseams.com/): Build the portal to launch your AI product. Who may call, which model they reach, and what they can spend: one path.
- [Pricing](https://ourseams.com/pricing): Same gateway, two cost models. Bring your own provider key and pay a fraction of a percent on spend we meter but never buy. Choose managed inference and the rate covers the tokens we buy for you.
- [Privacy Policy](https://ourseams.com/privacy): What is collected and what deliberately is not: no prompt or completion content anywhere, and no field to put it in.
- [Terms of Service](https://ourseams.com/terms): What the service does, what it costs, and who is responsible for what, including provider accounts under BYOK.
## Read instead of crawling
- [https://ourseams.com/llms-full.txt](https://ourseams.com/llms-full.txt): every page above, one file
- [https://docs.ourseams.com/llms.txt](https://docs.ourseams.com/llms.txt): documentation index
- [https://blog.ourseams.com/llms.txt](https://blog.ourseams.com/llms.txt): blog index
- [https://api.ourseams.com/v1/openapi.json](https://api.ourseams.com/v1/openapi.json): the seams API, applications, keys, models, routers, portals, and billing
## Where a reasonable assumption is wrong
- A stream can end with finish_reason "spend_limit_exceeded" and a usage block. It is not an error. The status was already 200 and cannot be retracted.
- Most caps never abort. When the remaining budget can be turned into a token count we clamp max_tokens instead, and the finish_reason is the ordinary "length".
- Model ids are "provider/model". A bare model name is not valid.
- Model names are per-bundle, so the same name can resolve to different models in different bundles.
- Spend is hard-capped per key. When the balance runs out the stream aborts; there is no soft overage.
- Gateway hosts are per application: https://<application-code>.gw.ourseams.com. A bare gw.ourseams.com host does not name an application.
## What we do not do
- Invoicing, revenue recognition, dunning and tax. Stripe does those; we do not wrap them.
- Merchant of record. Your customers pay you, and the legal relationship stays yours.
- Embeddings, images and audio. The gateway meters text and agent traffic.
- Your prompts and completions. There is no column for a message body and no code path that writes one.