Jev for classification
Jev, from TypeSafe AI, answers typed questions instead of writing text: pick one of these options, yes or no, a score in a range. For classification that makes it far cheaper and faster than a chat model, and every answer comes with a calibrated probability.
The video loads from YouTube only when you press play.
- 0.00002 USDto sort a support ticket into five categories with urgency and priority, in our tests
- 0.2 sto answer that ticket
- 76% vs 46%classification items fully right, Jev against the small chat model auto used before
- 91.6%of fields right across our classification test set
Two ways to use it
- Name it: send model jev-1.13.0 with response_format and a JSON schema whose fields are fixed choices, yes or no, or bounded numbers.
- Let auto pick it: the same request sent to auto goes to Jev whenever it is the cheaper answer, and to a chat model otherwise.
What Jev can answer
- A string from a fixed list of options, such as a category.
- A yes or no, such as urgent or not.
- A number within bounds, such as a priority from 1 to 5.
- Not free text, nested objects or lists. Send those to a chat model, or let auto decide.
Good jobs for Jev
- Support ticket triage: category, urgency and priority.
- Intent of an email or a chat message.
- Sentiment and topic of a review.
- Content moderation flags.
- Language detection and lead qualification.
What you get back
- The JSON your schema asked for, as the assistant message.
- A calibrated probability for each field in zabio.decision, so you can act on confident answers and review the rest.
- Under auto, a low-confidence answer is discarded and a chat model answers instead, and we pay for the discarded Jev call.
Questions
- 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.
Other features
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.
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.
Start free, upgrade when it's real
Change two lines in your code, send model auto, and every response shows what it cost and what you saved.