Pricing pages that hide the price
You should not need a spreadsheet and three tabs to know what a call will cost.
Model catalog
Every model you need, priced in plain sight
Filter the catalog by capability, sort by price, and see exactly what a model costs per token before you send it a single prompt. Switching is one field in your request.
Prices up front
per-token pricing per model
Filter by capability
chat, code, vision, reasoning and more
Switch in one field
change model_id, keep your code
The problem
What is slowing you down today
Locked into one model
A better or cheaper model ships and switching means a rewrite.
Guessing what a model can do
Capabilities are buried in docs, so you find out by trial and error.
Capabilities
Everything you need, nothing you don't
Capability filters
See which models handle chat, code, vision, reasoning and the other capabilities in the catalog.
Sort by price
Compare input and output token prices side by side.
Named or routed
Pin a model with model_id, or ask for a capability and let the router choose the lowest-priced fit.
Default model setting
Choose an account default in settings, or leave it unset and let the router decide per request.
Fail-over across the catalog
When a routed request's first choice is busy, another eligible model picks it up.
Try before you integrate
Every catalog model is one click away in the playground.
Use cases
What teams run on it
01
Cost tuning
Move routine traffic to lower-priced models without code changes.
02
Quality bake-offs
Run the same prompt against several models in the playground.
03
Capability expansion
Add vision or code features by changing the capability you request.
04
Consistency pinning
Lock a sensitive workflow to one named model.
How it works
From sign-up to production
- 1
Browse the catalog
Filter by capability and sort by price in the console.
- 2
Test in the playground
Compare outputs, tokens, latency and price.
- 3
Choose your strategy
Name a model_id, set a default, or route by capability.
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
How do I know what a model costs?
The catalog lists per-token input and output prices for every model, and each response reports its exact price.
Can I change models later?
Yes. Change the model_id or capability in your request. The endpoint and response shape stay the same.
What if I do not name a model?
Your account default is used if you set one. Otherwise the router picks the lowest-priced enabled model with the requested capability, which is chat if you do not specify.
Are models ever unavailable?
A model can be disabled in the catalog. Routed requests simply use another eligible model; a request naming a disabled model returns a clear error.
Find your model in a minute
Browse the catalog, compare prices, and try any model free.