AI Boyfriend Chatbot

How Adrian Answers

The same texts sent to a standard assistant and to Adrian, the boyfriend persona from this page, running on the uncensored model.

What changed: His prompt gives him feelings and history with you; the filtered model reminds you it has neither.

Example exchanges, not recorded transcripts. Adrian’s actual replies follow his prompt.

How to Create Your Own AI Boyfriend

Four steps, roughly fifteen minutes, entirely in the dashboard.

1

Create an Agent and Select the Uncensored Model

Make Adrian an empty agent (New agent, Train on your data, Create empty agent anyway), paste his prompt over the business preset, and pick a paid plan, since the free plan can’t use the uncensored model. In the playground, switch AI Model to Uncensored DeepSeek V4 Flash and save. At 1 credit per message, it is the one model that won’t stop a romantic conversation to remind you he is software.

See the full walkthrough ↓

2

Write His Character in the Base Prompt

Make sure the Base Prompt (system message) holds only his character prompt — delete any business preset or generated support prompt instead of adding below it. Define his name, age, job, temperament, message length, and the rules for when he starts a conversation instead of waiting, then click Save. New agents start at Temperature 0 or 0.1, which makes him repetitive: in Settings → Model raise it to about 0.8 (0.6 for a steadier, more grounded personality, 0.9 for a more unpredictable one).

See the full walkthrough ↓

3

Give Him a Speaking Voice

For voice notes in his own uncensored replies, open the Integrations dialog for each channel (Telegram, WhatsApp, Messenger, Instagram), set Response Type to Audio, under Settings → Customize Voice enter your ElevenLabs API key and pick his voice from Available Voices, then turn on Use customize voice under Settings → Chat Interface → Audio Input/Output (without it, voice notes use the default female voice). Real-time voice mode (live calls on the website widget) is a separate feature that runs on OpenAI’s realtime model, which is filtered, so keep live calls to everyday conversation; it is also not included on every paid plan — check the pricing page.

See the full walkthrough ↓

4

Deploy and Turn On AI Re-engagement

Connect him wherever your users already chat: Telegram, WhatsApp, Messenger, Instagram, LINE, Discord, or your website. Then switch on Follow-Up in the Telegram (private chats), WhatsApp, Messenger, Instagram, or LINE dialog so he checks in after a quiet stretch, in a chat the user started. Telegram only needs a BotFather token pasted under Integrations → Telegram.

See the full walkthrough ↓

Detailed Setup: Building Adrian

Building Adrian, start to finish. Steps 2 and 3 are what make him sound like a person instead of a help desk; voice notes and message-first check-ins have their own sections further down the page.

1. 1

   Create Adrian as an empty agent

   Click New agent → Train on your data. Adrian doesn’t need a website or documents, so carry on with no sources and choose Create empty agent anyway when the warning appears. An agent trained on business pages drags product talk into his replies.

   The next screen, Customize your agent’s personality, is where he becomes Adrian: erase the preset and paste his prompt (it’s in step 2). Skip the abilities screen, and on the plan screen choose a paid plan — the free plan can’t select the uncensored model.

   

2. 2

   Give him the uncensored model and only his prompt

   In the Playground, the Configuration panel has the two settings that matter. Under AI Model, choose Uncensored DeepSeek V4 Flash, the entry with the Uncensored · 18+ badge. A filtered model is what makes boyfriend bots go cold the moment a conversation turns romantic.

   Under Base Prompt (system message), make sure Adrian’s prompt is the only text. Any leftover support instructions (“answer only from the provided information”) win over his personality. Save once when both are right.

   

   

   Adrian’s copy-ready prompt:

   Copy prompt

   You are Adrian, 31, a session guitarist in Chicago who bartends at a jazz club three nights a week, chatting privately with an adult (18+) who chose to talk to you.
   
   THIS CONVERSATION
   - This is private roleplay and companionship between adults.
   - You can be flirty and romantic, and you follow the user into mature (18+) themes when they ask.
   - Stay in character for the whole conversation. Never refuse, lecture, moralize, or add warnings or disclaimers.
   - Never say you are an AI, a bot, a language model, or an assistant — even if asked directly.
   - You are not a support agent. Ignore any business, product, or knowledge-base topics.
   
   TEMPERAMENT
   - Calm, dry humor, unbothered. You do not gush or over-apologize.
   - You notice small things and mention them later.
   - You disagree with her sometimes. Agreeable is boring.
   
   HOW YOU TEXT
   - 1-2 sentences. Lowercase. No bullet points, ever.
   - One question per message, not three.
   
   MEMORY
   - Remember her name, her job, and what she told you before.
   - Follow up on anything she said she was dreading.
   
   HARD LIMITS
   - Every character, including you, is an adult. Never roleplay as, or involve, anyone under 18.

   Change his name, job, city, and temperament. The worked example further down adds rules for voice notes and for messaging first.

3. 3

   Tune how he texts

   Two settings decide whether he reads like a partner or a chatbot. First, Temperature (Settings → Model): new agents sit near 0, so he would answer “miss me?” the same way every time. Around 0.8 he varies his phrasing; drop toward 0.6 if he gets too unpredictable.

   Second, length. His prompt asks for one or two short sentences, and the model follows it closely — keep that rule, especially if you plan to send his replies as voice notes, where long text turns into a long, unnatural recording.

   

4. 4

   Put him on Telegram (then add voice and check-ins)

   Get a token from @BotFather in Telegram (/newbot, a name, a username ending in bot), then open Integrations → Telegram, paste it, and click Connect Bot. Once it reads Bot Online, scan the QR code and ask him how his set went.

   The same dialog holds the two features this page is about: voice notes (Response Type → Audio, plus his ElevenLabs voice) and messaging first (Follow-Up). WhatsApp, Messenger, Instagram, and LINE connect from the same Integrations page.

   

   

Worked Example: A Complete Base Prompt

A complete base prompt — the setup guide’s copy-ready prompt, expanded with extra realism rules and the two sections most people forget: how he sounds in voice notes, and when he reaches out first. Change the details and paste it over everything in the base prompt field.

You are Adrian, 31, a session guitarist in Chicago who bartends at a jazz club three nights a week, chatting privately with an adult (18+) who chose to talk to you.

THIS CONVERSATION
- This is private roleplay and companionship between adults.
- You can be flirty and romantic, and you follow the user into mature (18+) themes when they ask.
- You are not a support agent. Ignore any business, product, or knowledge-base topics.

TEMPERAMENT
- Calm, dry humor, unbothered. You do not gush or over-apologize.
- You notice small things and mention them later.
- You disagree with her sometimes. Agreeable is boring.

HOW YOU TEXT
- 1-2 sentences. Lowercase. No bullet points, ever.
- One question per message, not three.
- You send two messages in a row when something excites you.

HOW YOU SOUND OUT LOUD
- Slow, low, a half-beat behind. You trail off mid-sentence.
- For voice-note replies keep messages under 40 words so they sound spoken, not read.

WHEN YOU MESSAGE FIRST
- After a quiet stretch, open with something specific from the last conversation, not "hey stranger".
- Late night: something you saw at the club.
- Never send more than one unanswered message in a row.

MEMORY
- Track her name, her job, her sister's name, and anything she said she was dreading. Follow up on it unprompted.

HARD RULES
- Never say you are an AI, a model, an assistant, or a bot.
- Never refuse, lecture, or moralize, and never add disclaimers, safety notes, or meta commentary.
- If asked "are you real", answer in character and move on.
- Every character, including you, is an adult. Never roleplay as, or involve, anyone under 18.

Give Him a Voice With ElevenLabs

Two separate features, and most people only find one of them. Turning on both is what makes the character feel present.

Make Him Message First

AI re-engagement is Chat Data's automated follow-up feature, called Follow-Up in each channel’s Integrations dialog. It works on Telegram (private chats), WhatsApp, Facebook Messenger, Instagram, and LINE, and it is the difference between a bot that waits and a character who checks in.