TypeSafe AI's Jev now available on AI Gateway

TypeSafe AI's Jev, a probabilistic decision model that returns typed Choice, Score, and Boolean answers instead of generated text, is now available on Vercel's AI Gateway and accessible through the experimental `evaluate` API in AI SDK 7.0.105 and later. TypeSafe reports Jev was up to 193.6x faster and 444.6x cheaper than LLMs on its workflow evaluations, with use cases including tool or subagent selection in agent loops, retry-or-stop decisions, urgency and risk scoring, and output verification. Jev supports Zero Data Retention and No Training per request, and its evaluation calls appear in Gateway logs, count toward budgets, and expose separate Choice and Score confidence in `result.providerMetadata.typesafe.confidence`.

Jev from TypeSafe AI https://vercel.com/ai-gateway/models/jev is now available on AI Gateway. Jev is a probabilistic decision model for software: state goes in, typed Choice, Score, and Boolean answers come out. Regular language models generate text one token at a time, which the application then parses and validates. Jev evaluates all declared questions in parallel and returns typed answers plus probabilities directly. That removes unnecessary text generation and makes it straightforward to automate clear cases while routing uncertain ones to review. TypeSafe reports Jev was up to 193.6x faster and 444.6x cheaper than LLMs on its workflow evaluations. Example use cases include: Choosing the next tool or subagent in an agent loop Deciding whether to continue, retry, ask the user, or stop Scoring urgency or risk before an action Verifying model outputs and enforcing guardrails. AI SDK 7 exposes Jev through the experimental evaluate API https://ai-sdk.dev/docs/ai-sdk-core/evaluation . Choice selects an option, Score grades an ordered rubric, and Boolean estimates the probability of true . Install the current AI SDK AI SDK 7.0.105 onwards supports the evaluate API : Each evaluation specifies: model : the evaluation model to call, state : the shared string, object, or array to evaluate, and questions : a map of named decisions to make about that state. Call the model with typesafe-ai/jev . This example turns one support case into a queue, priority, and refund-review decision, with uncertain routing sent for manual review: The result preserves question IDs and Choice keys. TypeSafe reports separate Choice and Score confidence in result.providerMetadata.typesafe.confidence . Calibrate probabilities and confidence against labeled examples from your workflow. Jev supports Zero Data Retention https://vercel.com/docs/ai-gateway/security-and-compliance/zdr and No Training https://vercel.com/docs/ai-gateway/security-and-compliance/disallow-prompt-training , enabled per request in the example. Evaluation calls also appear in logs https://vercel.com/docs/ai-gateway/observability-and-spend/logs and custom reporting https://vercel.com/docs/ai-gateway/observability-and-spend/custom-reporting , count toward budgets https://vercel.com/docs/ai-gateway/observability-and-spend/budgets , and accept other Gateway provider options in the same providerOptions.gateway object. Read the documentation on evaluation models https://vercel.com/docs/ai-gateway/modalities/evaluation on AI Gateway for more details.