# thinQit Dex

> An EU-hosted decision API. Typed answers with calibrated probabilities, the same answer every time.

Dex answers typed questions (pick, rate, check) about a text or JSON state with calibrated probabilities. Identical requests to the same GPU model version return identical answers. Processing is in the EU; request content is not stored by default. You pay for input tokens only. Status: sign-up and API keys are open (test keys are free); paid top-ups open soon; SDK packages are coming soon to PyPI and npm, with interim downloads at https://thinqit.ai/docs/reference/sdks/#downloads. API base URL: https://api.thinqit.ai. Contract: https://thinqit.ai/openapi.yaml.

## Docs

- [Dex documentation](https://thinqit.ai/docs/index.md): What the decision API does, the three question types, how a call looks, and where to read next.
- [Quickstart](https://thinqit.ai/docs/quickstart.md): Get a key, try the API against a local mock today, make a call in curl, Python or TypeScript, and act on the answer.
- [State](https://thinqit.ai/docs/concepts/state.md): The material every question is about, sent once per request as a string or as JSON, with its limits and billing.
- [Questions](https://thinqit.ai/docs/concepts/questions.md): The question object, question ids, the pick, rate and check types, criteria, and how questions in one request relate.
- [Field references](https://thinqit.ai/docs/concepts/field-references.md): Point a question at one field of a JSON state with the path syntax, without paying for that field twice.
- [Confidence](https://thinqit.ai/docs/concepts/confidence.md): How the confidence value of each answer type is computed, what it does not promise, and how numbers are rounded.
- [Abstention](https://thinqit.ai/docs/concepts/abstention.md): Let an unsure answer say so with min_confidence, and route it to a person instead of acting on a guess.
- [Determinism](https://thinqit.ai/docs/concepts/determinism.md): The same request bytes to the same GPU model version return the same answers, byte for byte. Scope, limits, method and checks.
- [Calibration](https://thinqit.ai/docs/concepts/calibration.md): What calibrated probabilities mean, how each model version is calibrated and measured, the measured numbers, and the bars they are held to.
- [Models and versions](https://thinqit.ai/docs/concepts/models.md): The model alias, exact versions and what they freeze, the fallback versions, the version lifecycle and the models endpoint.
- [Fallback and served_by](https://thinqit.ai/docs/concepts/fallback.md): When a hosted fallback in the EU answers instead of the GPU, how every response says so, and how to keep a request on the GPU.
- [Known limits](https://thinqit.ai/docs/concepts/known-limits.md): Where Dex answers are weaker today, with the measured numbers, and what to do about each.
- [Authentication and API keys](https://thinqit.ai/docs/reference/authentication.md): Send your API key as a bearer token, check which key you use, and validate the key format and checksum offline.
- [Errors](https://thinqit.ai/docs/reference/errors.md): The error envelope, every status code with its type and codes, which errors to retry, refunds, and the SDK error classes.
- [Headers](https://thinqit.ai/docs/reference/headers.md): The request and response headers of the API, including request ids, idempotency and the rate limit headers.
- [Rate limits](https://thinqit.ai/docs/reference/rate-limits.md): Request, token and concurrency limits per key and per account, how they are counted, and what to do on a 429.
- [Idempotency](https://thinqit.ai/docs/reference/idempotency.md): Retry a request safely with an idempotency key, so a network failure never makes you pay twice.
- [Limits](https://thinqit.ai/docs/reference/limits.md): Every size limit of a decide request, the error each one returns, and the model limits from the models endpoint.
- [Tokens and billing](https://thinqit.ai/docs/reference/billing.md): Output is free and you pay for input tokens only, EUR 0.05 per million. What counts, how a charge is computed and the prepaid balance.
- [Data handling and residency](https://thinqit.ai/docs/reference/data-handling.md): Where requests are processed, what is stored and for how long, the fallback exception, roles under the GDPR and security basics.
- [SDKs and CLI](https://thinqit.ai/docs/reference/sdks.md): The Python and TypeScript SDKs, the command line tool and the VS Code extension, with downloads, configuration, retries, idempotency and errors.
- [Changelog](https://thinqit.ai/docs/changelog.md): Changes to the website, the documentation, the API contract and the model versions, newest first.

## API reference

- [API reference overview](https://thinqit.ai/docs/api/index.md): every route, grouped by tag
- [Answer typed questions about a state](https://thinqit.ai/docs/api/decide.md): POST /v1/decide
- [List model aliases and exact versions](https://thinqit.ai/docs/api/listModels.md): GET /v1/models
- [The API key making this call](https://thinqit.ai/docs/api/getCurrentKey.md): GET /v1/key
- [Usage for one API key over a date range](https://thinqit.ai/docs/api/getUsage.md): GET /v1/usage
- [Current prepaid balance and price tier](https://thinqit.ai/docs/api/getBalance.md): GET /v1/balance
- [Email a one-time sign-in code](https://thinqit.ai/docs/api/requestSignInCode.md): POST /v1/console/auth/email-code (console route)
- [Exchange a sign-in code for a session](https://thinqit.ai/docs/api/createSession.md): POST /v1/console/auth/session (console route)
- [Sign out](https://thinqit.ai/docs/api/deleteSession.md): DELETE /v1/console/auth/session (console route)
- [Account profile, balance and settings](https://thinqit.ai/docs/api/getAccount.md): GET /v1/console/account (console route)
- [List API keys](https://thinqit.ai/docs/api/listKeys.md): GET /v1/console/keys (console route)
- [Create an API key](https://thinqit.ai/docs/api/createKey.md): POST /v1/console/keys (console route)
- [Read one API key](https://thinqit.ai/docs/api/getKey.md): GET /v1/console/keys/{key_id} (console route)
- [Rename a key, change its scopes or lower its limits](https://thinqit.ai/docs/api/updateKey.md): PATCH /v1/console/keys/{key_id} (console route)
- [Revoke an API key](https://thinqit.ai/docs/api/revokeKey.md): DELETE /v1/console/keys/{key_id} (console route)
- [Start a Stripe Checkout top-up](https://thinqit.ai/docs/api/createCheckoutSession.md): POST /v1/console/checkout-sessions (console route)
- [Account usage as a time series](https://thinqit.ai/docs/api/getUsageSeries.md): GET /v1/console/usage/series (console route)
- [Run a decision from the console in test mode](https://thinqit.ai/docs/api/playgroundDecide.md): POST /v1/console/playground/decide (console route)

## Optional

- [Home and recorded races](https://thinqit.ai/index.md): what Dex is, how it works, and Dex against LLMs on four recorded workflows
- [Use cases](https://thinqit.ai/use-cases.md): when to use Dex instead of an LLM, ten business cases and the cases it does not suit
- [Pricing](https://thinqit.ai/pricing.md): EUR 0.05 per million input tokens, output free
- [Comparison, on the pricing page](https://thinqit.ai/pricing.md): Dex is an EU-hosted alternative to Jev (TypeSafe AI). A sourced comparison is on the pricing page. (Source: https://docs.typesafe.ai/models, read 2026-09-25)
- [Legal](https://thinqit.ai/legal/): Draft outlines of the terms, acceptable use policy, privacy policy, data processing agreement and subprocessor list.
- [Terms of service](https://thinqit.ai/legal/terms/): Draft outline of the terms for using the decision API, its console, SDKs and command line tool.
- [Acceptable use policy](https://thinqit.ai/legal/acceptable-use/): Draft outline of what the decision API may not be used for, including the EU AI Act and GDPR limits.
- [Privacy policy](https://thinqit.ai/legal/privacy/): Draft outline of how account, billing and request metadata are handled, and for how long they are kept.
- [Data processing agreement](https://thinqit.ai/legal/dpa/): Draft outline of the agreement under which request content is processed on the customer's behalf.
- [Subprocessors](https://thinqit.ai/legal/subprocessors/): Draft list of the third parties that help run the decision API, what each one does, and whether it sees request content.
- [Everything in one file](https://thinqit.ai/llms-full.txt)