FAQ
Getting started
Why should I use SIRAYA Model Router?
SIRAYA Model Router provides a unified API to access all the major LLM models on the market. It also allows users to aggregate their billing in one place and keep track of all of their usage using our analytics.
SIRAYA Model Router bills each model at its published list price, while pooling the uptime of the underlying providers, with a unified API and fallbacks so that you get much better uptime.
What makes SIRAYA Model Router unique?
SIRAYA Model Router stands out as a unified routing layer that connects multiple AI model providers through a single, consistent API. Instead of integrating separately with different LLM or embedding services, developers can use SIRAYA Model Router to simplify model management, request routing, and version control.
SIRAYA Model Router offers flexible configuration options—such as automatic provider selection, fallback routing, and performance optimization—which help ensure reliability and cost-efficiency. In short, SIRAYA Model Router makes it easier to build and scale AI applications by abstracting away provider complexity while maintaining full transparency and control.
What's the story behind SIRAYA Model Router?
SIRAYA Model Router was created to solve a growing pain in the AI development world: managing multiple model providers efficiently. As the ecosystem of large language models and embeddings expanded, developers often found themselves juggling different APIs, authentication methods, and data formats for each provider. This added unnecessary friction and slowed down innovation.
Seeing this challenge, the creators of SIRAYA Model Router envisioned a single, unified routing layer that could abstract away these complexities—allowing developers to focus on what matters most: building great products powered by AI. The idea was to give teams the flexibility to mix and match providers, experiment seamlessly, and improve reliability through smart routing and fallbacks.
From that vision, SIRAYA Model Router emerged as an infrastructure solution designed to make multi‑provider AI development as simple, scalable, and transparent as possible. It reflects the broader effort to move from fragmented model integrations toward a cohesive, provider‑agnostic AI ecosystem.
Why should a person choose SIRAYA Model Router over its competitors?
SIRAYA Model Router offers a flexible and developer‑friendly way to manage multiple AI model providers through one unified API. Unlike tools that tie you to a single vendor, SIRAYA Model Router lets you easily switch or combine models from different sources without changing your application code.
SIRAYA Model Router provides built‑in routing logic, fallback mechanisms, and usage tracking so you can optimize cost, latency, and reliability automatically. In addition, its configuration‑based approach and detailed observability tools simplify scaling and debugging. In short, SIRAYA Model Router helps teams focus on building AI‑powered features rather than maintaining complex provider integrations.
Who are the primary audience of SIRAYA Model Router?
The primary audience of SIRAYA Model Router includes developers, product teams, and organizations building applications that rely on AI models or large language models (LLMs).
SIRAYA Model Router is designed for engineers who need to integrate, manage, and optimize access to multiple AI providers without maintaining separate APIs. Startups, enterprise AI teams, and platform builders can all benefit from its unified routing system—especially those seeking flexibility, scalability, and cost control in multi‑provider environments. In essence, SIRAYA Model Router serves anyone who wants to simplify AI infrastructure while maintaining high performance and reliability.
How do I get started with SIRAYA Model Router?
To get started, create an account and add credits on the Credits page. Credits are simply deposits on SIRAYA Model Router that you use for LLM inference. When you use the API or chat interface, we deduct the request cost from your credits. Each model and provider has a different price per million tokens.
Once you have credits you can create API keys and start using the API. You can read our quickstart guide for code samples and more.
How do I get support?
Contact us through the contact page.
How do I get billed for my usage on SIRAYA Model Router?
For each model we have the pricing displayed per million tokens. There is usually a different price for prompt and completion tokens. There are also models that charge per request, for images and for reasoning tokens. All of these details will be visible on the Logs page.
When you make a request to SIRAYA Model Router, we receive the total number of tokens processed by the provider. We then calculate the corresponding cost and deduct it from your credits. You can review your complete usage history in the Activities page.
Chat responses include token counts in usage, which you can combine with the model's list prices to estimate the cost. Responses do not include a cost field.
Any discount is agreed individually for your account through SIRAYA sales and shows up as the discounted price in your billing logs. There is no public discount schedule.
Pricing
What are the prices for using SIRAYA Model Router?
Each model is billed at its list price, as published on the Models page. Any discount is agreed individually for your account through SIRAYA sales; there is no public discount schedule.
For more details on our model price, please see our Models page.
For more details about every request cost, please see our Logs page.
How is the billing calculated when Prompt Cache is enabled?
Input tokens read from the cache are billed at the model's cache-read price; the remaining input tokens are billed at the normal input price. Claude models also charge a cache-write price when cache_control creates a new cache entry. See the Models page for per-model prices.
Models and Providers
What LLM models does SIRAYA Model Router support?
SIRAYA Model Router provides access to a wide variety of LLM models, including frontier models from major AI labs.
For a complete list of models you can visit the Models page or fetch the list through the models api.
How frequently are new models added?
We work on adding models as quickly as we can. We often have partnerships with the labs releasing models and can release models as soon as they are available.
I am an inference provider, how can I get listed on SIRAYA Model Router?
Reach us through the contact page.
How does model fallback work if a provider is unavailable?
If a provider returns an error SIRAYA Model Router will automatically fall back to the next provider. This happens transparently to the user and allows production apps to be much more resilient.
API Technical Specifications
What authentication methods are supported?
SIRAYA Model Router authenticates requests with API keys, passed as Bearer tokens in the Authorization header (Authorization: Bearer <API_KEY>). See Authentication.
What API endpoints are available?
SIRAYA Model Router exposes OpenAI-compatible and Anthropic-compatible endpoints, so you can call any model with the same request/response format:
- Text:
/v1/chat/completions,/v1/responses, and/v1/messages(Anthropic format) - Images:
/v1/images/generationsand/v1/images/edits - Video:
/v1/videos/generations, plusGET /v1/videos/{video_id}for status andGET /v1/videos/{video_id}/contentfor the result - Embeddings:
/v1/embeddings - Models:
/v1/models
See the API Reference for details.
What are the supported formats?
The API supports text, images, and PDFs. Images can be passed as URLs or base64 encoded images. PDFs are sent through Chat Completions on models that accept file input; see that page for the supported formats.
How does streaming work?
Streaming uses server-sent events (SSE) for real-time token delivery.
Set stream: true in your request to enable streaming responses.
What SDK support is available?
SIRAYA Model Router is a drop-in replacement for OpenAI. Therefore, any SDKs that support OpenAI by default also support SIRAYA Model Router. Check out our docs for more details.
Can I mix different modalities in one request?
Yes. You can send text and images — and PDFs on models that accept file input — in the same request, and the model processes all inputs together. Supported input types vary by model; check the Models page.
Does SIRAYA Model Router use Prompt Cache by default?
It depends on the model provider. Most providers (OpenAI, Google, xAI, DeepSeek, and others) cache automatically with no configuration (implicit caching). Claude models only cache when you mark content with cache_control (explicit caching). See Prompt Caching.
Will using Prompt Cache change my token usage or latency?
- Output: Caching only reuses processed input tokens. It does not change the output content.
- Latency: Requests that hit the cache are generally faster, especially for long prompts.
- Billing: Input tokens read from the cache are billed at the model's cache-read price, which is usually much lower than the normal input price. See the Models page for per-model prices.
Can I disable Prompt Cache?
Implicit caching is managed by the model provider and cannot be turned off. For Claude models, caching only happens when you add cache_control, so leave it out if you don't want caching.
Privacy and Data Logging
Please see our Terms of Service and Privacy Policy.
What data is logged during API use?
We log basic request metadata (timestamps, model used, token counts). Prompt and completion are not logged by default. We do zero logging of your prompts/completions, even if an error occurs.
Credit and Billing Systems
What purchase options exist?
SIRAYA Model Router uses a credit system where the base currency is US dollars.
All of the pricing on our site and API is denoted in dollars. Users can top up their balance manually.
My credits haven't showed up in my account
If you paid using Stripe, sometimes there is an issue with the Stripe integration and credits can get delayed in showing up on your account. Please allow up to one hour. If your credits still have not appeared after an hour, contact us and we will look into it.
How to monitor credit usage?
The Activity page allows users to view their historic usage and filter the usage by model, provider and api key.
We also provide a Logs page that has live information about the balance and remaining credits for the account.
How do volume discounts work?
Discounts are agreed individually for each account; there is no public volume-discount schedule. If you expect high volume, contact us.
What payment methods are accepted?
We accept all major credit cards, AliPay, PayPal, WechatPay and Invoice.
Account Management
What analytics are available?
Our Activity dashboard provides real-time usage metrics. If you would like any specific reports or metrics please contact us.
Input Format Support
What about video support?
Video generation is available through the Video Generation API (text-to-video and image-to-video). Some chat models also accept video as input — check the Models page for each model's supported input types.