Every model is another integration
New auth, new error formats, new invoices. Your roadmap stalls while engineers babysit plumbing.
For SaaS teams
Ship the AI feature on this sprint
Smart compose, summaries, search answers, onboarding copilots. Add them to your product behind one endpoint, with per-call pricing you can map straight onto your own plans.
Live in minutes
from API key to first feature call
Clear unit cost
price per call, visible before you ship
One integration
swap models without a redeploy
The problem
What is slowing you down today
Feature cost is a guess
Product wants to price the AI tier. Finance wants a unit cost. Nobody can say what a summary actually costs.
One outage, one angry status page
When the single model behind your feature has a bad hour, your customers see it first.
Capabilities
Everything you need, nothing you don't
One endpoint for every feature
Summaries, drafts, classification and code help all go through the same authenticated POST, so new features reuse what you already built.
Per-call unit economics
Every response carries token counts and price, so you can attribute cost to a tenant, a feature or a plan.
Prices before you call
The model catalog shows per-token pricing for each model, so product and finance can agree on a feature before it ships.
Fail-over your users never notice
If one model is busy, the request moves to another that can serve it. Your feature stays responsive.
Cache for the common questions
Popular prompts repeated inside the cache window come back fast and cost nothing.
Keys per environment
Create separate keys for staging and production and revoke one without touching the others.
Use cases
What teams run on it
01
In-app summaries
Turn long threads, tickets or documents into a crisp recap right where your users work.
02
Smart compose
Draft emails, notes and replies inside your product in the user's own context.
03
Natural-language search answers
Pair your search results with a short generated answer that cites what you found.
04
Onboarding copilots
Answer setup questions in-product so new accounts reach value faster.
How it works
From sign-up to production
- 1
Prototype in the playground
Try prompts against catalog models and see tokens, latency and price per run.
- 2
Wire one endpoint
Call POST /api/ainf/inference from your backend with a bearer key. Existing chat-style clients adapt with a base URL change and a small body mapping.
- 3
Ship and measure
Track daily and month-to-date usage in the console as the feature rolls out.
Control
Built so you stay in charge
Keys you control
API keys are shown once at creation and stored only as a hash. Revoke any key instantly from Settings.
Spend you can see
Every request carries its own receipt: tokens in, tokens out, latency and price. Nothing is billed that you cannot trace.
Limits you set
Per-account rate limits and a free monthly allowance keep experiments safe before they become production.
Data you decide on
You choose exactly what you send. Repeat prompts can be served from your cache, and cache hits are never billed.
For developers
One endpoint. Every model.
Send a prompt and a capability, or name a model. One bearer key, one small JSON body, and a full receipt on every response.
curl https://YOUR-DOMAIN/api/ainf/inference \
-H "Authorization: Bearer $AI_INFERENCE_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: order-1042" \
-d '{"prompt":"Summarise this ticket in one line","capability":"chat"}'Questions, answered
Can I bill my own customers for AI usage?
Every response includes tokens and price, and your keys are tracked per request, so you have the data to build usage-based pricing on your side.
Do I need to rewrite my existing client?
Usually not. Point it at our base URL, send your key as a bearer token and put your text in the prompt field.
How do I keep staging traffic separate?
Create a key per environment, and tag each response's price and tokens in your own logs so staging and production spend stay separate.
How am I billed?
Metered per token at the price shown for each model. Usage beyond the free monthly allowance is invoiced every 30 days and paid online.
Your next AI feature is one endpoint away
Prototype today on the free allowance, then ship with pricing you can explain.