Zabio vs Portkey
Portkey, now part of Palo Alto Networks, is an AI gateway built for observability, guardrails and governance in large companies, working with your own provider keys. Zabio is built to lower the bill: it picks the model for each request, gives you eight vendors from one prepaid balance, and takes no cut of what it saves you.
At a glance
| Zabio | Portkey | |
|---|---|---|
| What it is | A hosted router that picks the model for you | An AI gateway for observability and security, now part of Palo Alto Networks |
| How it charges | Provider list price plus a (see the pricing page) service fee, and (see the pricing page) per auto request for automatic model choice. No cut of what routing saves, no subscription, no fee on top-ups | Free developer tier and paid plans for the hosted platform. You pay the providers directly |
| Its fee is based on | The provider price of each call, plus a fixed fee per auto request. Never your saving | Your plan and logged requests |
| Automatic model choice | Per request, the model and its reasoning setting, from public benchmark results by task and difficulty, on three lines you can name, with extra care on high-stakes jobs, and a request split across models when that is cheaper | Routing rules you write: conditions, load balancing and fallbacks |
| What you see per call | The model, why it was chosen, the cost and the saving, in the response itself | Detailed logs, traces and analytics |
| Your own provider keys | No service fee on those calls; auto requests pay (see the pricing page) each for automatic model choice | Always your own keys |
| Teams and spend limits | Unlimited organizations, member quotas, per-key caps and model allow lists, and an audit log, all included | Budgets, rate limits and access control; advanced governance on higher plans |
| Models | 114 chat models from 10 vendors | 1,600+ models |
| Where it runs | Hosted by Zabio in the United States | Hosted, or the open-source gateway on your servers |
Where Portkey beats us
- You need deep observability across your whole AI stack: detailed logs, traces and analytics.
- You need guardrails and governance for a large organization.
- It is part of Palo Alto Networks, a far larger company, with a security business behind the product.
- You want to run the gateway yourself. The Portkey gateway is open source.
Where we beat Portkey
We care most about one number: how much we save you. Everything below follows from that.
- You want the gateway to lower your bill on its own: routing picks the cheapest model that does the job, sizes reasoning to it, and splits requests when that is cheaper.
- You want one prepaid balance for eight vendors, with no account to open at each.
- You want pricing you can check in your head: a (see the pricing page) service fee and (see the pricing page) per auto request, no plans or tiers, and no cut of what routing saves.
- You want the cost and the saving of every call in the response itself.
Switching from Portkey
Portkey and Zabio both speak the OpenAI API. Point base_url at https://api.zabio.ai/v1, use your Zabio key, and send model auto.
What makes the difference, in depth
Smart Splitting
One request, several models, a smaller bill, decided for you on every call. When we can save, we save.
Smart routing
The right model and reasoning setting for every request, picked from evidence, with the reason on every response.
Jev for classification
Tagging, routing, moderation and sentiment for a fraction of the price of a chat model, named or picked by auto.
Caching
Your history cached for you, the 1 hour cache when turns are slow, and never a cut of your cache discount.
Flex
Half price when you can wait, with one deadline field.
Batching
Thousands of requests in one file at about half price, with auto routing on every line.
Teams and keys
Organizations, member quotas, keys with spend caps and model lists, and your own vendor keys with no service fee.
Questions
- How is Zabio different from Portkey?
- Portkey, now part of Palo Alto Networks, is an AI gateway focused on observability, guardrails and governance for large companies, and it works with your own provider keys. Zabio focuses on lowering the bill: it picks the model for each request, gives you eight vendors from one prepaid balance, and charges a small service fee per call plus a small fixed fee per auto request, with no cut of what it saves you.
- What is Smart Splitting?
- Under auto, a request that carries a long document and asks about only part of it can be split: a cheap model copies out the lines the question needs, word for word, Jev checks that the extract holds everything the answer needs, and the model auto picked answers from that extract instead of the whole document. If the check fails, Zabio sends the full document and pays for the attempt itself, so a split never costs you more. In our tests it cut the cost of long-document questions by about a third on average, and by more than half on contracts, with every fact kept. It never applies to a model you name, every step shows in zabio.split, and zabio.split set to off turns it off for a request.
- Can Zabio classify text for less than a chat model costs?
- Yes. Name jev-1.13.0, the decision model from TypeSafe AI, and send a JSON schema whose fields are fixed choices, yes or no answers, or bounded scores. Zabio turns each field into one of the typed questions Jev answers and returns the JSON, with a calibrated probability for each field in zabio.decision. Sorting a support ticket into one of five categories with an urgency flag and a 1 to 5 priority cost about 0.00002 USD and took about 0.2 seconds in our tests. Through auto, Zabio uses Jev on its own for such requests whenever it is the cheaper answer; in our tests it got 76 percent of classification items fully right, against 46 percent for the small chat model auto picked before, and answered about five times faster.
Checked on 28 September 2026 from each company's public pages. Prices and features change. If something here is out of date, tell us at hello@zabio.ai and we will fix it. Sources: Portkey.