{"slug": "ai-gateway-now-supports-typesafe-clients-and-an-http-api-for-jev", "title": "AI Gateway now supports TypeSafe clients and an HTTP API for Jev", "summary": "Vercel's AI Gateway added support for TypeSafe AI's Jev probabilistic decision model via existing TypeSafe clients and a new HTTP API, in addition to the AI SDK. Jev takes state as input and returns typed answers with probabilities attached, offering three question types — boolean (probability from 0 to 1), choice (one option from a named set), and score (rating against a defined scale) — and is named as typesafe-ai/jev in new integrations. Requests are billed through AI Gateway on all three paths, and Jev serves as the default evaluation model for eve, Vercel's framework for building durable agents.", "body_md": "You can now call [Jev from TypeSafe AI](https://vercel.com/ai-gateway/models/jev) through AI Gateway using an existing TypeSafe client or the HTTP API, in addition to the AI SDK.\n\n**TypeSafe client:** Point an existing TypeSafe client at AI Gateway without changing its evaluation calls.\n\n**HTTP API:** Call Jev directly from any language or framework.\n\n**AI SDK:** Call Jev from a TypeScript application using AI SDK.\n\nJev is a probabilistic decision model for software. State goes in, and typed answers come out with probabilities attached, so there's no generated text to parse. Requests are billed through AI Gateway on all three paths, so they appear alongside your other model calls in usage and observability.\n\nChange the base URL and API key. Your `systemOne` calls, `noul` questions, and response shapes stay exactly as they are.\n\nNew integrations name the model as `typesafe-ai/jev` and ask one of three question types: `boolean` returns a probability from 0 to 1, `choice` picks one option from a set you name, and `score` rates against a scale you define. This example asks whether an agent should keep working after fixing a bug and passing its tests.\n\nWith the HTTP API, `POST` to `/v1/evaluate`:\n\nWith the AI SDK, run the same evaluation through `evaluate`:\n\nYou can also use Jev through [eve](https://vercel.com/eve), a framework for building and deploying durable agents with sandboxed compute, human approvals, and evaluations already built in. eve uses Jev as the default [evaluation model](https://eve.dev/docs/guides/evaluate) for automatic model selection, typed evaluations, and automated tool approvals.\n\nSee the [TypeSafe-compatible API guide](https://vercel.com/docs/ai-gateway/sdks-and-apis/typesafe) to migrate a client, or the [evaluation guide](https://vercel.com/docs/ai-gateway/modalities/evaluation) for the HTTP API or AI SDK.", "url": "https://wpnews.pro/news/ai-gateway-now-supports-typesafe-clients-and-an-http-api-for-jev", "canonical_source": "https://vercel.com/changelog/ai-gateway-now-supports-typesafe-clients-and-http-api-for-jev", "published_at": "2026-09-21 00:00:00+00:00", "updated_at": "2026-09-21 22:55:04.689367+00:00", "lang": "en", "topics": ["ai-products", "ai-agents", "developer-tools", "ai-infrastructure"], "entities": ["Vercel", "AI Gateway", "TypeSafe AI", "Jev", "AI SDK", "eve"], "alternates": {"html": "https://wpnews.pro/news/ai-gateway-now-supports-typesafe-clients-and-an-http-api-for-jev", "markdown": "https://wpnews.pro/news/ai-gateway-now-supports-typesafe-clients-and-an-http-api-for-jev.md", "text": "https://wpnews.pro/news/ai-gateway-now-supports-typesafe-clients-and-an-http-api-for-jev.txt", "jsonld": "https://wpnews.pro/news/ai-gateway-now-supports-typesafe-clients-and-an-http-api-for-jev.jsonld"}}