Skip to main content
Adapt-1 is currently the most direct API access to Core. Observations and measured outcomes update its learned representations, rules, predictive models, and policies. Applications use the retained state through structured predictions, decisions, evidence, or control commands. REI Chat brings continuing reasoning and state into a conversational interface. Use it to interact through language or add conversational responses to an application.

Adapt-1

Build systems that learn and reason during operation, from causal discovery and adaptive decisions to continuous control.

REI Chat

Set up an agent and connect it through the chat API.

​
Start building

Make an Adapt-1 call

Authenticate and get your first structured result.

Start with REI Chat

Authenticate and make your first conversational API call.

Use cases

Find a workflow that connects your inputs, results, and measured outcomes.

Get a first learned result

Send observations and query the state they form.

​
Research and resources

Further reading

Explore technical write-ups and evaluation reports on adaptive state, online learning, Discovery, and continuous control.

Research direction

Read the research goals for Core.

Lab and resources

Browse research artifacts, model releases, and supporting material.