Jev-compatible API

Use Decision Models as a Jev-compatible API

Decision Models is independent and offers a Jev-compatible API: you send a state and typed questions, and you get typed answers with probabilities. Jev is a trademark of TypeSafe AI; Decision Models is not affiliated with TypeSafe AI.

Get API keyQuickstartPricing

What “Jev-compatible” means here

Decision Models accepts the documented Jev request shape, state plus typed questions (noul, choice or score). Migration is not drop-in: you use your own Decision Models API key and a Decision Models model ID, and processing tiers, billing, limits and errors are our own. See the Jev alternatives comparison and the migration notes for the differences.

Base URL and request

The base URL is https://api.decisionmodels.io; text requests go to POST /v1/systemone. Send your key as Authorization: Bearer, and choose the processing region with the S1-Region header (eu or global, subject to key permissions).

curl
curl --fail-with-body --silent --show-error --max-time 15 \
  "${DECISIONMODELS_BASE_URL:-https://api.decisionmodels.io}/v1/systemone" \
  -H "Authorization: Bearer $DECISIONMODELS_API_KEY" \
  -H 'Content-Type: application/json' \
  -H 'S1-Region: eu' \
  --data '{"model":"s1-fast","state":"Mia owns a red bicycle.","questions":{"color":{"type":"choice","instructions":"Which color is the bicycle?","criteria":{"red":null,"blue":null}}}}'

The same example is tested in the migration checklist, which also has Python, Node.js, LiteLLM and Ollama examples. GET /v1/models returns the live model list.

Sign in, get a key, top up

To start, sign in with an email link or Google, create a key in the dashboard and redeem an invite code or top up from EUR 1. Details are in the quickstart; rates are on the pricing page.

Models offered

Models from our models page; the full current list is there and in the live registry, GET /v1/models.

Model IDInputDescription
s1-fasttextbased on Blink v0.3 26B-A4B (Apache-2.0), NVFP4
s1-protextmammoth-1 27B by baselabs (MIT), FP8 on our own engine in the EU tier (s1-pro-5); Peer-to-Peer tier: s1-pro-3 (vLLM-SR d3)
s1-visiontext + imagebased on Blink v0.3 26B-A4B (Apache-2.0), NVFP4; one image plus text
s1-llm-auto-routertexts1-llm-auto-router (CPU routing classifier)

Already using a Jev client?

The client switch guide lists the base URL setting for each supported SDK and client. The migration checklist walks through key, model ID, limits and error handling before you switch traffic.

Docs and comparisons