The conversation engine every agent runs on

Streaming replies, long-term memory, fourteen languages, images and structured output β€” one brain, shared by every surface.

12 capabilities Β· 9 on by default

What it is

This is the part people mean when they say "the agent". It answers, it remembers, it produces something β€” a plan, a quote, a shortlist β€” rather than only talking. Everything else on the platform is either feeding this module or reading what it produced.

How it works

  1. 1

    One brain, every surface

    The hosted page, the embed widget, the API and the channels all call the same engine, so behaviour cannot drift between them.

  2. 2

    Memory runs alongside the reply

    Facts are extracted in parallel with the answer and compressed into a durable profile, so returning visitors are recognised instead of re-interviewed.

  3. 3

    Structured output, not just prose

    The same turn that answers can emit typed records β€” a task, a lead, an event β€” which is how the dashboard fills itself.

Everything in this module

Each one is a switch on your agent. Nothing here is an upsell.

  • Conversational agent

    The core AI, streaming replies.

  • Multilingual

    Answers in the visitor’s language.

  • Structured generation

    The agent’s core deliverable (plan/config/output).

  • Long-term memory

    Remembers across sessions.

  • Welcome-back

    Continuity for returning users.

  • Voice modeoff by default

    Speak + read aloud.

  • Image analysisoff by default

    Upload a photo for analysis.

  • Image generation & editing

    Create and retouch visuals.

  • Locked characters

    Save a character once; every later image and video is the same person.

  • Image packs

    Social sets, comic/BD pages and campaign visuals from one anchor.

  • Video generationoff by default

    Anchor a still, animate it, then face-lock. Needs a video provider key.

  • Conversation history

    Saved & reloaded.

Before it runs: Voice mode, image analysis and video generation ship off. Voice and image analysis are a toggle; video needs a provider key of your own β€” the image providers that ship enabled do not generate video.

Who this is for

Everyone. There is no configuration in which this module is off β€” the rest of the platform exists to give it something to work with.

Questions

What is included in the ai chat & assistant module?

12 capabilities, 9 of them on by default: conversational agent, multilingual, structured generation, long-term memory, welcome-back and more.

Do I have to pay extra for this module?

No. Every agent is built on the whole platform β€” modules are not sold separately. What varies by plan is volume: how many agents and how many conversations, not which capabilities you are allowed.

Can I turn parts of it off?

Yes. Each capability is a switch on the agent, and switching one off removes it from the agent’s behaviour as well as from its interface β€” the agent stops offering what it cannot do.

Which parts are not on out of the box?

Voice mode, Image analysis, Video generation. Voice mode, image analysis and video generation ship off. Voice and image analysis are a toggle; video needs a provider key of your own β€” the image providers that ship enabled do not generate video.

Every agent gets this module

Describe what you do in one sentence. The agent is built on the whole platform β€” then you switch off whatever you do not want.

Build your agent free

Agents

More

πŸͺ
Find Providers
πŸ‘€
Profile
❓
FAQ