OpenRouter lets you call Jev, TypeSafe’s decision model, through the same OpenAI-compatible API it uses for other models, plus a separate alpha endpoint built for typed decisions. OpenRouter’s own provider page lists pricing for two Jev model ids, though a discrepancy with OpenRouter’s public models feed means that pricing is not independently confirmed.
What the integration does
OpenRouter is a hosted layer in front of many AI model providers, exposed through one shared API key. TypeSafe has added Jev to that list, so requests that would otherwise go straight to TypeSafe’s API can instead route through OpenRouter.
TypeSafe describes Jev as a decision model, though most of what it does is best understood as a classification model: it sorts input into labels you supply at request time, with a probability on each one.
Two paths exist. The standard OpenAI-compatible API at https://openrouter.ai/api/v1 lists Jev as a model id like any other. An alpha “Decisions” endpoint, which OpenRouter calls its “Decisions router,” is built around Jev’s typed output and suits jobs like intent and model routing, where you want a probability on a fixed set of options rather than generated text.
For calling Jev from Vercel or Cloudflare instead, see the separate Vercel and Cloudflare integration pages.
Official support
OpenRouter’s own provider page, at openrouter.ai/typesafe, lists Jev with pricing. That listing comes from OpenRouter itself, not from a third party. The page describes Jev as “a structured decision model from TypeSafe, and the first of its System One models,” optimized for routing, classification, and decisions, returning typed choices rather than free text.
This is where the page conflicts with OpenRouter’s own data. A check of OpenRouter’s public model-listing API, GET https://openrouter.ai/api/v1/models, on 2026-09-20 returned zero matches for “typesafe” or “jev,” even though the provider page shows two priced models. Nothing found while researching this page reconciles that gap, so treat the pricing below as what the provider page states, not as something OpenRouter’s models API confirms.
| Model id | Context window | Input price | Output price |
|---|---|---|---|
typesafe/jev-latest |
32K tokens | not shown | not shown |
typesafe/jev-1.13 |
32K tokens | $0.042 per million tokens | $0 per million tokens |
typesafe/jev-latest always redirects to the newest Jev model. The provider page also shows a running usage counter, “342B tokens processed” as of 2026-09-20, which climbs over time and is not a fixed spec.
Setup
- Get an OpenRouter API key. Access to Jev depends on OpenRouter’s account permissions, not on TypeSafe.
- To call Jev through OpenRouter’s standard API, send requests to
https://openrouter.ai/api/v1withmodelset totypesafe/jev-1.13ortypesafe/jev-latest, in the normal OpenAI-compatible format, authenticated with your OpenRouter key. - To call Jev through the alpha Decisions endpoint instead, send a
POSTrequest tohttps://openrouter.ai/api/alpha/decisionsfortypesafe/jev-latest, OpenRouter’s “Decisions router.” It is in alpha and OpenRouter has published no dedicated documentation page for it as of 2026-09-20. - To route TypeSafe’s own SDKs through OpenRouter instead, point them at
POST https://openrouter.ai/api/v1/systemone, matching TypeSafe’s direct API request and response shape, and set your OpenRouter key asTYPESAFE_API_KEYor pass it through the SDK’sapiKeyoption. - For a worked example of that shape, see how to get Jev access, since it matches TypeSafe’s own API rather than introducing anything new.
What people built with Jev and OpenRouter
The examples linked below are projects and write-ups that used Jev specifically through OpenRouter, rather than through TypeSafe’s own API directly.
Limits and gotchas
Rate limits for Jev access through OpenRouter are not published as of 2026-09-20. Do not assume parity with TypeSafe’s direct API.
typesafe/jev-latest always points at the newest Jev release, so a request that works today can behave differently once TypeSafe ships a new version. Pin typesafe/jev-1.13 if you need a fixed target.
The pricing above, $0.042 per million input tokens and $0 per million output tokens for typesafe/jev-1.13 as of 2026-09-20, comes from OpenRouter’s provider page alone, since OpenRouter’s public models API does not list Jev. Compare it with TypeSafe’s own numbers on the speed and pricing guide before relying on it.
The “342B tokens processed” counter on the provider page is a running total, not a fixed benchmark, and it will read differently by the time you check it.
The Decisions endpoint is alpha software. Treat it as less stable than the standard API until OpenRouter documents it properly.
FAQ
Does OpenRouter support Jev officially?
OpenRouter’s own provider page at openrouter.ai/typesafe lists Jev with pricing, so yes in that sense. But OpenRouter’s public models API, GET https://openrouter.ai/api/v1/models, returned zero results for “typesafe” or “jev” when checked on 2026-09-20. The two OpenRouter sources disagree, and that gap is unresolved as of this writing.
Do I need a TypeSafe API key to use Jev on OpenRouter?
No, an OpenRouter API key is enough on its own. itsmostafa/typesafe-mcp’s README confirms that an OPENROUTER_API_KEY can substitute for TYPESAFE_API_KEY if that OpenRouter account has access to Jev. You only need a direct TypeSafe key if you are calling TypeSafe’s API without OpenRouter in the path at all.
What does Jev decide when called through OpenRouter?
Same thing it decides anywhere else, since OpenRouter does not change the model itself. OpenRouter’s provider page describes Jev as a decision model that returns typed choices instead of free text, built for tasks like routing and classification. Calling it through OpenRouter’s standard API or the alpha Decisions endpoint only changes how the request reaches the model. What the model decides stays the same.
Is it free to use Jev on OpenRouter?
Not entirely. OpenRouter’s provider page lists typesafe/jev-1.13 output at $0 per million tokens, but input is priced at $0.042 per million tokens as of 2026-09-20, per that same page. typesafe/jev-latest shows no price at all on the listing. None of this pricing is confirmed by OpenRouter’s own public models API, which does not list Jev.
Examples for OpenRouter
AIBackends
An AI API server that abstracts multiple models and providers behind one layer, for local use with Ollama or LM Studio or in the cloud via OpenRouter, OpenAI, Anthropic or Google. It now supports Jev.
jev-eval-agent
Personal-assistant agent built with Vercel's eve framework and 100 mocked tools, served through OpenRouter. Measures how many steps the agent needs when the LLM picks the tool itself versus when Jev picks it via a 101-option Choice.
Jev 1.13 on OpenRouter
Model page for Jev 1.13 on OpenRouter, reachable through OpenRouter's OpenAI-compatible endpoint. Listed at $0.042/M input, $0/M output, 32K context.
Jev on OpenRouter (TypeSafe provider page)
OpenRouter lists two TypeSafe models, typesafe/jev-latest and typesafe/jev-1.13, both with a 32K context window, at $0.042/M input and $0/M output.