DISCOVERY · ENTERPRISE
Syniray Enterprise
The leading models, through an API you already know.
A pay-as-you-go API for companies and independent developers: 14 AI models behind one OpenAI-compatible endpoint, at the providers' list prices.
Or write to [email protected]
Early access
The platform and the developer documentation are in early access and open gradually. If you cannot get in yet, write to [email protected].
Overview
Build with the models, pay for what you use
Syniray Enterprise is for building AI into your own products and tools. You create an account — for a business or for yourself — load a balance, create API keys and call the models. There is no subscription: each request is paid from your balance.
It is a separate product from the Syniray plans, with its own account, keys and billing.
The API
One endpoint, many models
Chat completions
An OpenAI-compatible chat completions endpoint, with answers streamed as they are written.
Model list with prices
An endpoint that lists the models available to you, with their prices and their reasoning levels.
Usage and balance
Read your usage and your current balance from the API.
Reasoning when you ask
Reasoning is turned on per request, with a reasoning effort each model accepts.
Clear errors
Errors in the familiar OpenAI format: an invalid key, an insufficient balance, an unavailable model, too many requests.
Pricing
List prices, one transparent fee
Provider list prices
Tokens are charged at the providers' published prices, with no markup per token.
A fee on top-ups
An 8% processing fee is added when you load your balance — you receive exactly the amount you chose.
Prepaid balance
Load from $50 by card, or turn on automatic top-up below a threshold you set.
Invoices
Every top-up has its invoice, with your tax ID when you add one.
The platform
Everything you need to run it
Overview
Balance, monthly spend, and charts of requests, tokens, cost, latency and errors, with filters.
API keys
Create, disable and revoke keys by project. A key is shown in full only once.
Usage
Tables and charts by day and month — by model, by project and by key.
Logs
Each request with its time, model, tokens, cost and status. The content of requests is never stored.
Projects
Separate your work into projects, each with its own keys.
Teams
For a company or for one person
Business accounts
Invite your team and give each person a role:
- Owner
- Admin
- Developer
- Viewer
- Billing
Personal accounts
For freelancers and independent developers — the same API and platform, without the company setup.
Security
Designed so a key can do as little as possible
Management only in the platform
Keys, billing and members are managed only from a signed-in session. An API key can call the models — never create keys or change billing.
Keys stay on your servers
The API is not meant to be called from a web page: it sends no cross-origin headers, so a key belongs on your server.
No request content stored
Logs keep what is needed to account for usage — never the prompts or the answers.
Today
What the API does not do yet
Single sign-on (SSO/SAML), webhooks, fine-tuning, embeddings, image and audio models, regional hosting and balances in currencies other than US dollars are not part of Syniray Enterprise today.
Use cases
Where teams use it
AI in your product
Add chat and text generation to your own application.
Internal tools
Assistants and automations for your team, on the models you choose.
Many models, one integration
Switch between models from different companies without new integrations.
Questions
Frequently asked questions
Is Syniray Enterprise the same as a Syniray subscription?
No. The Syniray plans are subscriptions for using Syniray, Code and the Academy. Enterprise is an API for building your own products, paid per use from a prepaid balance.
How is it priced?
Tokens at the providers' list prices, with no markup per token. An 8% processing fee is added when you load your balance.
Which models are available?
14 models from the Syniray catalogue. Syniray 2.0 is not offered through the API.
Does it work with OpenAI clients?
The chat completions endpoint follows the OpenAI format, so a client that speaks that format can use it by pointing to the Syniray endpoint with a Syniray key.
Do you store my prompts?
No. Logs keep the time, model, tokens, cost and status of each request — never its content.
Can my team share one account?
Yes, with a business account: invite members and give each one a role — owner, admin, developer, viewer or billing.
Do you offer SSO or an SLA?
Not today. Single sign-on and a service level agreement are not part of Syniray Enterprise yet.
How do I start?
Open the platform at https://platform.syniray.ai/ to create your account, and read the documentation at https://developers.syniray.ai/. Both are in early access — to ask for access, write to [email protected].
Build with Syniray Enterprise
Open the platform to create your account and API keys, or start with the documentation.
Or write to [email protected]