TypeSafe logo

Jev 1.14

TypeSafeReleased Oct 14, 2026typesafe/jev-1.14.0

Jev 1.14 is the newest release of TypeSafe's flagship System One model, served by POST /v1/systemone. It evaluates a `state` (text sent as a string, JSON object, or array of text values, plus PNG and JPEG images anywhere in the request) against a map of typed `questions` and returns one calibrated `answer` per question. Supported question types are choice, score, noul (yes/no), and box, with box questions requiring Jev 1.14. The context limit is 64k tokens per request and 32k per question, rising to 262k for both with the TypeSafe-Danger-Zone-Long-Context header. Trained with RLCD for calibrated decisions; not fine-tuned on customer data or trained on customer requests. Charged per input token, output tokens are free.

Context
64K
Input /1M
$0.04
Output /1M
—
AcceptsTextImageProducesDecisionsTokenizer Jev

Providers

FastRouter routes your requests to this provider. You can also bring your own key.

Typesafe AI logo
Typesafe AI
typesafe
Input /1M$0.04
Output /1M—
Context64K
Max output—

Supported parameters

Request parameters you can send with this model.

Parameter
Type
Typesafe AI
Other
questions
object
state
string
-

Make your first API call

OpenAI-compatible. Point your SDK at FastRouter, or call the API directly.

Full API docs
Replace <FASTROUTER_API_KEY> with your key · Get your key →
HTTP

Frequently asked questions

More models from TypeSafe

TypeSafe logo
Jev 1.13typesafe/jev-1.13.0

Jev 1.13 is the previous release of TypeSafe's flagship System One model, served by POST /v1/systemone. It evaluates a `state` (text sent as a string, JSON object, or array of text values, plus PNG and JPEG images anywhere in the request) against a map of typed `questions` and returns one calibrated `answer` per question. Supported question types are choice, score, and noul (yes/no); box questions need Jev 1.14. The context limit is 64k tokens per request and 32k per question, rising to 262k for both with the TypeSafe-Danger-Zone-Long-Context header. Trained with RLCD for calibrated decisions; not fine-tuned on customer data or trained on customer requests. Charged per input token, output tokens are free. Pin this ID if you have tuned confidence thresholds against it, since the `jev-latest` alias has moved to Jev 1.14.

TypeSafe logo
Jev (latest)typesafe/jev-latest

Jev is TypeSafe's flagship System One model, served by POST /v1/systemone. It evaluates a `state` (text sent as a string, JSON object, or array of text values, plus PNG and JPEG images anywhere in the request) against a map of typed `questions` (choice, score, noul yes/no, and box) and returns one calibrated `answer` per question. `jev-latest` is an alias for the most recent stable release and currently resolves to `jev-1.14.0`; it moves when a new release ships, so pin a versioned ID if you have tuned confidence thresholds. The context limit is 64k tokens per request and 32k per question, rising to 262k for both with the TypeSafe-Danger-Zone-Long-Context header. Audio and video input are not supported. Trained with RLCD for calibrated decisions; not fine-tuned on customer data or trained on customer requests. Charged per input token, output tokens are free.

Model scores sourced from ArtificialAnalysis