Help center

What thinkhx lets you do, the dashboard you land on, and the quickest path to your first result — run a ready-made app or build your own.

Create an application, configure its model, data and instructions, activate it, and manage its name, logo, visibility and deletion.

Browse and search the public applications that are ready to run, and open one to get a result right away.

Attach an AI model to your application with a short guided wizard, change it later, or remove it — the model decides what your app can do.

Understand the models thinkhx offers — how they're grouped by task, what the Text, Image, context-size and size badges mean, and which ones your plan unlocks.

Run an application — fill in its inputs, send the request, follow its progress, and read the result.

Browse every reasoning your project has run, inspect a run's inputs and response, search for a specific one, and review the runs of a single app.

Read an application's run totals and token usage, open its run history, and start a new run — from the Overview tab of the run console.

See your plan and its monthly usage allowance, change plan, buy a one-off token pack, pay by card or another method, and stop an automatic renewal.

Find the invoices for your project's plan renewals and packs, read what each column means, and download an invoice as a PDF.

Track how many reasonings your project runs over time and how many tokens they consume, see finished, new and error runs per day, and find your heaviest applications and models.

Create an API key to connect external systems to your project, copy it once at creation, then rename, disable or delete it.

The base URL, the Bearer key, the scope grammar and the JSON envelope every thinkhx endpoint returns, with a complete curl example you can run straight away.

Every HTTP status the thinkhx API returns, the authentication and scope failures that produce them, and the exact messages to match on.

Create, list, read, edit and delete the applications of your project, including their input schema, their instructions and their public or private visibility.

Browse the public apps published by every project, read one of them with its input schema, and use the keyless endpoint that powers a public showcase.

Launch a reasoning against an app, poll it until it finishes, read its inputs, response and token counts, and list a project's run history.

Read the project's plan, the billing cycle it runs on, the tokens consumed and still available, and the lifetime totals of finished, failed and pending runs.

Read the project your key belongs to and update its name, description, default language, company details, postal address and icon colours.

List the members of your project, read one of them, invite a consumer by email, change a member's role and remove a member.

Read the account your key acts as, including its email, phone, security and login activity, and update its name, surname and language.

List a project's API keys, inspect the one you are calling with, create a key with a scope and an expiry, and revoke a key.

List the file categories and the files of your project, read one file with its delivery URL and dimensions, and delete a file.

Read the project's audit trail, with the type of each change, the record it touched, where it came from and the address it was made from.

Send feedback from your project and read back the messages it has already sent, with their topic and status.

Read the project's plan and its totals for team members, API keys, files and events in a single call.

Whether you need an account, why everything you see belongs to one project and how to look at another, which actions need administrator rights, and where your plan and remaining allowance are shown.

What a token is, which runs consume it, when your allowance resets, why creating an application or using a model can be refused, what happens when the allowance runs out, and how plan changes, renewals, consumption dates and invoices behave.

Why an application cannot be run or published, what a reasoning's status and origin mean, what to do when a run fails, how long your reasonings are kept, and what happens to them when you delete an application.