Pricing that goes down when your bill does
You pay the provider list price for every call, plus a small service fee, and a small fixed fee when Zabio picks the model. We take no cut of what routing saves: every cent is yours. No subscription, no seats, no fee on top-ups.
Start free, pay when it's real
Every account starts on Free. Top up when your prototype becomes a product.
Free, for real
100 requests a day on six models from OpenAI, Google, Mistral and Together AI. auto routing included. No card, no expiry.
Pay as you go
Top up once and everything opens up: every model, no daily limits, the same simple pricing.
| Free | Pay as you go | |
|---|---|---|
| Price | 0 USD, forever | Provider list price plus a (see the pricing page) service fee, and (see the pricing page) per auto request |
| Card | Not needed | Top up from (see the pricing page) |
| Models | 6 models from OpenAI, Google, Mistral and Together AI | Every model, from every vendor |
| Requests per day | 100 | No daily limit |
| Requests per minute | 20 | Higher, set per key |
| Size per request | 16,000 tokens in, 2,000 out | Each model's full context and output |
| auto, cheap, balanced, best | Among the free models | Among every model |
| Smart Splitting, Jev, caching | Included | Included |
| Batch, flex and background jobs | Included, each line counts as a request | Included |
| Cost and saving on every call | Included | Included |
| No training on your data | Included | Included |
| Team members | 1 | Your whole team |
| API keys | 2 | As many as you need, with spending caps per key |
| Fair use | A daily fair-use limit | None |
Twice OpenRouter's free requests, on the vendors' own models. Zabio pays the vendors, so your prompts travel on their paid API terms: no training on your data.
Start free. Top up when your prototype becomes a product.
Pricing you can check in your head
Zabio adds a small service fee to every call, plus a small fixed fee when it picks the model for you. We take no cut of what routing saves: every cent of the saving stays in your pocket.
billed = provider_price
+ (see the pricing page) of provider_price
+ (see the pricing page) per auto request
saved = baseline_price - billed, all yoursIn the worked example below, routing moves one request from a 0.0500 USD model to a 0.0100 USD one. You pay (see the pricing page) and keep (see the pricing page).
A fee that goes up is announced at least 30 days ahead, by email to every organization owner. The numbers on this page come from the pricing plan in force.
What you save, and what Zabio charges
Every call is billed at the provider list price plus Zabio's fees, each on its own line. We take no cut of what routing saves: every cent of the saving is yours.
- A service fee of (see the pricing page) on the provider price of every call. A call to a model you name pays this fee and (see the pricing page).
- On every auto request, a fixed (see the pricing page) for automatic model choice, paid once per request whatever happens inside it.
A worked example
You send a short request with model auto. The classifier finds an easy job and the balanced line takes a cheap model. Your key's baseline model over the last 30 days is GPT-5.5.
| Provider list price of the call | 0.0100 USD |
|---|---|
| Service fee on the provider price | (see the pricing page) |
| Automatic model choice, once per request | (see the pricing page) |
| Billed | (see the pricing page) |
| Baseline: the same tokens on your key's baseline model | 0.0500 USD |
| You saved | (see the pricing page) |
The whole saving is yours: Zabio's fees are the same whether routing saved a lot or nothing. Your baseline is the model your key used at the 95th percentile of price over the last 30 days: the most expensive 5 percent of calls are set aside, so one burst does not raise it. It is priced with the same cache hits your call had, so a cache discount never counts as routing's saving. The classifier that reads the job is part of the fixed fee and is never billed on its own. Calls served with your own vendor keys pay no service fee, and their auto requests pay (see the pricing page).
The same request, routed and not routed
| Routed with auto | Baseline model, named directly | |
|---|---|---|
| Provider list price | 0.0100 USD | 0.0500 USD |
| Service fee | (see the pricing page) | (see the pricing page) |
| Fixed fee for the request | (see the pricing page) | (see the pricing page) |
| You pay | (see the pricing page) | (see the pricing page) |
How we measure what we saved you
Every response shows what the call would have cost on your usual model, so you can see what routing saved. These rules keep that number honest.
- Your baseline is the model your key actually uses: the 95th percentile of the prices of its calls over the last 30 days, so one expensive burst does not raise it.
- It is priced with the same cache hits your call had, so a cache discount never counts as a saving.
- A new key needs 20 calls before its baseline is set, usually within its first hour. Until then its calls are their own baseline and show no saving, and the dashboard shows how many calls are left.
- When you name a model, that model is the baseline: the call saves nothing, and Zabio's fees show as a small negative saving.
Credits and spend controls
Prepaid, in US dollars. No subscription. No fees on top-ups.
Free plan to start
Every account starts on Free: 100 requests a day, no card. Your first top-up unlocks every model and removes the limits.
Credits by card
Add credits through Stripe Checkout, from (see the pricing page) up to 5,000 USD per payment. No fee: 50 USD paid is 50 USD of credit.
Provider list price plus two small fees
A service fee of (see the pricing page) on the provider price of every call, and (see the pricing page) per auto request for automatic model choice. On your own vendor keys, no service fee and (see the pricing page) per auto request. We take no cut of what routing saves, or of your cache and batch discounts.
Spend controls
Monthly caps per API key, monthly quotas per member, requests per minute and tokens per minute limits per key.
Two ways to earn Zabio credit
Credit you earn joins your balance, works on any call and never expires.
We match your top-ups, up to 250 USD
Spend 250 USD in the 30 days after you sign up, and from then on we match every top-up dollar for dollar until we have added 250 USD. Add 20, we add 20. Add 50, we add 50.
25 USD for every developer you bring
Share your link. When someone who signs up through it adds 100 USD, you get 25 USD in credit once their top-ups have cleared 30 days. Every referral counts.
Offer credit has no cash value. A top-up that earns a match can't be refunded, and a disputed one loses its match. Read the terms
The fine print, in plain words
Automatic model choice
Every auto request pays one fixed fee, (see the pricing page), for reading the job and picking the model, whatever happens inside it: a reused classification, Smart Splitting's two calls, a move up a line or an answer by Jev. The classifier itself is never billed to you.
Smart Splitting
We pay for deciding whether to split a request, and for any split that falls back to the full document. Splitting can only lower your bill.
Flex
Send a deadline of 15 minutes or more and calls can run on the flex tier of OpenAI or Google at about half the price. The saving is measured against the flex price of your baseline, so the discount stays yours.
Bring your own vendor keys
You can give Zabio your own API key for any vendor it serves. Calls to that vendor then run on your key and the vendor bills you.
- Free except (see the pricing page) per auto call. Calls served with your own key pay no service fee, and a call to a model you name on your key costs nothing through Zabio.
- You keep the routing, sessions, stored conversations and the dashboard.
- Keys are verified with one cheap call, encrypted with a key unique to your organization, and never shown again.
- If a vendor rejects your key, Zabio stops using it and tells you. It never falls back to its own key for that vendor.
Pricing questions
- What fees does Zabio charge?
- Two small ones, and never a cut of your saving. Every call pays a service fee, a percentage of the provider price, which covers what it costs us to serve it: infrastructure, payment processing and support. Every auto request also pays a small fixed fee for automatic model choice, once per request, which pays for reading the job and picking the model. Calls on your own vendor keys pay no service fee, and their auto requests pay the fixed fee. Every cent routing saves stays with you. The current numbers are on the pricing page, and a fee that goes up is announced at least 30 days ahead by email to every organization owner.
- Is Zabio free?
- Yes. Every account starts on the Free plan: 100 requests a day, up to 16,000 tokens in and 2,000 out per request, on six models from OpenAI, Google, Mistral and Together AI, with auto choosing among them. No card, no expiry. Every other model stays visible, one top-up away, and a daily fair-use limit keeps Free sustainable. Your first top-up removes every limit. Calls on your own vendor keys pay no service fee.
- How is Zabio different from other AI routers?
- It picks the model for you, to get you more right answers for less. Send auto and Zabio reads each request, picks the model and reasoning setting that fit the job, and shows why on every response, with what the call cost and what it saved. Pricing stays simple: the provider's list price, a small service fee on each call and a small fixed fee on each auto request. Zabio takes no cut of what routing saves, so every cent of the saving is yours.
- Can I control what each person and each key spends?
- Yes. Owners and admins set a monthly quota per member, and every API key can have its own monthly spend cap, request and token limits per minute, and a list of the models it may use. Calls stop with 429 quota_exceeded when a limit is reached.
- Can I pay less when I am not in a hurry?
- Yes. Send zabio.deadline with a duration of 15 minutes or more, for example "30m", and Zabio serves the call on the flex tier of OpenAI or Google where the model has one, at about half the price. Under auto, candidates are compared at their flex prices. If the vendor has no capacity, Zabio retries at the normal tier and bills the price actually used, so a flex request never fails for that reason. Your saving is measured against the flex price of your usual model, so it never counts the vendor's discount as routing's work. For large jobs that can wait hours, batching sends a whole file at the vendors' batch prices.
- Does Zabio support prompt caching?
- Yes, on by default where the vendor offers it. For Anthropic, Zabio places cache markers on your system prompt, tools and conversation history, so a long agent loop reads its history from the cache instead of paying for it again on every turn. OpenAI caches automatically. When the turns of a session are more than five minutes apart, Zabio switches to the 1 hour cache so the history is still there. Cached tokens are billed at the vendor's cached rate and shown on every response, and you can set a 5 minute or 1 hour cache per key or per call.
Your first 100 requests today are on us
Change two lines in your code, send model auto, and every response shows what it cost and what you saved.
Prices for every model are provider list prices in USD per million tokens. Read the billing page for the full rules.