Documentation
Salyro documentation
Salyro puts one OpenAI-compatible API in front of OpenAI, Anthropic, Google Gemini and Grok. Requests pass through to the provider that owns the model you name, and every credential, key, log and cost belongs to the gateway it was made on. Beside that API, Behaviors let you version a prompt, a model and its parameters in Salyro and call the result by name.
Get started
2 pagesCore concepts
4 pages- GatewaysThe unit of isolation the product is built around — what a gateway owns, what isolation means in practice, and when to use more than one.
- ProvidersConnecting your own OpenAI, Anthropic, Gemini and Grok accounts to a gateway, and what happens to a credential once it is saved.
- API keysThe lifecycle of a Salyro key — issuing it, the rate limits attached to it, and revoking and replacing it without downtime.
- BehaviorsA Behavior is a typed, versioned AI contract your application calls by a stable key — what one is made of, how a version becomes live, and how to run it.
Operate
3 pages- Logs & conversationsWhat is recorded for every request, how conversation history differs from a request log, and how long either is kept.
- Usage & costsHow usage is measured and cost is calculated from it, which breakdowns exist, and how the figure relates to a provider's invoice.
- ErrorsThe families a failure falls into, which of them are worth retrying, and which need something changed before the next attempt.
Reference
1 pageLooking for a whole task rather than one capability?
The pages above describe what each part of Salyro does, one capability at a time. The guides work through complete scenarios instead, from the first step to a working result.
Browse the guides