FAQ

Frequently asked questions

Short answers about the Decision Models API, taken from our pricing page, docs and legal pages. New to the idea? Start with What are decision models?

What is Decision Models?

Decision Models hosts low-cost decision models for routing, classification and scoring behind one Jev-compatible API, with Peer-to-Peer and EU tiers. You send a state and typed questions and get typed answers with probabilities.

Which models are available?

The models offered on the API today are s1-fast, s1-pro, s1-vision, s1-llm-auto-router. The live list is returned by GET /v1/models, and the models page describes each one. Our models are open-weights models; s1-pro uses mammoth-1 27B by baselabs (MIT) in the EU tier.

How is it priced?

Prices are per 1 million input tokens; output tokens are free. For accounts created on or after 12 November 2026 the list price of s1-fast is $0.032 (Peer-to-Peer) or $0.034 (EU), and of s1-pro $0.097 (Peer-to-Peer) or $0.103 (EU). USD rates exclude applicable taxes; EUR rates are on the pricing page.

What is the difference between the EU tier and Peer-to-Peer?

Peer-to-Peer starts on spare EU capacity, with EU requests taking priority. When busy, Peer-to-Peer can draw on rented worldwide GPU capacity from independent hardware providers, including outside the EU/EEA, but only for keys with “Allow worldwide processing” enabled. EU-tier requests always stay in the EU.

Are prompts or answers stored?

Our gateway does not log prompts or answers, and we do not use them for training. Independent providers on network-enabled keys may store content; use the EU tier for personal data, secrets or confidential content. For billing we keep usage records such as token counts.

How do I get an API key?

Sign in with a one-time email link or with Google, create an API key in the dashboard and send your first request. Free allowance by invitation. Top up from EUR 1. Private individuals and businesses worldwide can sign up and top up. Taxes are shown in checkout before you pay.

Is Decision Models the Jev service?

No. Decision Models is a separate service hosting open-weight decision models behind a Jev-compatible API. Jev is a trademark of TypeSafe AI; references are nominative only. Decision Models is not affiliated with TypeSafe AI.

Can I switch an existing Jev client?

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.

What are the request limits?

s1-pro accepts up to 3 text questions per request and s1-fast up to 1. The input limit per question, including its state, depends on the model and is 8,192 tokens for s1-pro and 16,192 for s1-fast. Too many questions return 400 request_limit_exceeded; exceeding the rate limit returns HTTP 429 rate_limit_exceeded with a Retry-After header.

Can I send images?

Yes, with s1-vision: one image per request (PNG, JPEG or WebP). Text-only requests are billed at the s1-vision text rate; a request that contains an image is billed at the s1-vision image rate for all its input tokens (text and image positions), because image processing takes about 3× the GPU time per token. An image counts as the input tokens the model actually reads.

Can I get a private, dedicated endpoint?

Yes. On the model pages of open-weights models, dedicated hosting gives you a private GPU endpoint that nobody else shares, with your own API keys. You can pick EU or Global.

How do I ask for a model that is not on the API yet?

On a model page for a model we do not offer yet you can ask for one email when it becomes available on the API. We send one email to confirm your address first (double opt-in), use your email address and the model only for that notice, and delete the entry once the notice is sent (at the latest 12 months after you sign up; unconfirmed entries after 30 days).

Jev is a trademark of TypeSafe AI. Decision Models is not affiliated with TypeSafe AI.