Question 1

Is Apertis compatible with the OpenAI SDK?

Accepted Answer

Yes, Apertis is fully compatible with the official OpenAI SDK. Simply change the base URL to api.apertis.ai and use your Apertis API key. Your existing code works without any modifications.

Question 2

Which AI providers does Apertis support?

Accepted Answer

Apertis routes requests across 30+ AI providers including OpenAI, Anthropic (Claude), Google (Gemini), AWS Bedrock, Azure OpenAI, Mistral, Cohere, and many more through one OpenAI-compatible control plane.

Question 3

How does automatic failover work?

Accepted Answer

When a provider experiences downtime or errors, Apertis automatically routes your requests to an equivalent backup model. This happens transparently with zero code changes required.

Question 4

How does pricing work?

Accepted Answer

Apertis uses a credit-based system. You pay for what you use with transparent per-token pricing. There are no minimum commitments or hidden fees. You can start with free credits and only pay as you scale.

Question 5

What is prompt caching and how does it save costs?

Accepted Answer

Prompt caching stores and reuses responses for identical or similar prompts. When the same prompt is sent again, Apertis returns the cached response instantly without calling the underlying provider - saving you both money and latency. Cache reads are free.

Question 6

Does Apertis have its own SDK?

Accepted Answer

Yes! Apertis offers official SDKs for JavaScript/TypeScript (npm i @apertis/ai-sdk-provider) and Python (pip install apertis). You can also use the standard OpenAI SDK by simply changing the base URL.

Question 7

Is my data secure with Apertis?

Accepted Answer

Yes. Apertis processes requests in memory and does not store your prompts or responses beyond what is needed for caching. We use TLS encryption for all data in transit. Your API keys are encrypted at rest.