What is TalkyTalky Ai?

TalkyTalky Ai manages 5 data types including bot profiles. Helps you organize, track, and share your work in 1 place for teams and solo users.

Frequently Asked Questions

Everything you need to know about TalkyTalky — organised by topic so you can jump straight to what you need.

🚀Getting Started & Onboarding

🤖Creating AI Partners — the Profiles Page

💬Chat — Layout & Tools

✍️Your First Chat — Text

🧠Advanced Chat — Memory & Context

🎙️Voices & LiveStream Mode

🧩LLMs — What They Are & How to Choose

⚡Advanced LLM Features

🔑BYOK — Bring Your Own Key

🛡️Safety & Mental Health

🔍Technical & Safety Deep Dive — The Founder's Audit

⚡New Features & Tools — Deep Research, Tasklet, Exa & Voice

Still have questions?

Can't find what you're looking for? Email us at support@talkytalky.ai or use the 🐞 Bug Report form in the footer.

Frequently Asked Questions

What is TalkyTalky?

How do I create an account and get started?

What do I get access to as a new user?

What are the Terms of Use and why do I need to accept them?

I signed up but nothing is loading — what do I do?

What is a 'Profile' and how is it different from a chat?

How do I create a new AI partner?

What is the Advanced Character Builder?

Can I assign a specific Large Language Model (LLM) to a partner?

Can I assign a specific voice to a partner?

What is the Guide (Eve / Adam)?

Can I have multiple AI partners?

What does the Chat screen look like?

What is the toolbar below the message input?

What are the icons in the chat header toolbar?

What is the Tone Switcher Indicator?

How do I pin a message to Memory?

How do I send my first message?

What should I say first?

Why does my partner seem to forget things from a previous session?

How do conversation titles work?

Can I start a new conversation with the same partner?

How does memory work in TalkyTalky?

Does memory work across different AI models — even when the Tone Switcher changes models mid-conversation?

What is the Memory page?

What is Semantic Memory Search?

How much context does my partner have per conversation?

Can I search my conversation history?

Can I export my conversations?

What voice options are available?

What is LiveStream mode and why is it so fast?

What is 'Click to Play' vs 'Auto' audio playback?

How do I set up a voice for my partner?

My partner isn't speaking — what's wrong?

Which voice provider is best for non-English languages?

What is an LLM?

What models does TalkyTalky use?

What is the difference between GPT, Claude, and Gemini?

Which model should I choose as a beginner?

What is 'Auto' model routing?

What AI models are available?

Can I use Grok 4.6?

What is Kimi K3?

How do I manually assign a model to a partner?

Can I assign a specific model to one partner and a different one to another?

What does the Tone Switcher Indicator show me?

What is the monthly budget feature on the BYOK page?

Can I use multiple LLM providers at the same time?

What is the Frontier Credit Power Meter?

How does the intelligent Tone Switcher decide which model to use?

What does BYOK mean?

Why would I use BYOK instead of the standard plan?

Which platforms support BYOK?

How do I connect a BYOK key?

Is my API key safe?

Should I rotate (replace) my API keys regularly?

Do I need a BYOK key to use TalkyTalky?

How much does BYOK cost?

Can I bring my own API keys?

Is TalkyTalky a replacement for therapy or professional support?

What happens if I mention self-harm or crisis situations?

Are my conversations private?

What content is not permitted?

How are accounts siloed for privacy and safety?

How do I report a concern or bug?

How does TalkyTalky handle image generation safety for minors?

Is my data being used to train the models?

What happens in a mental health crisis?

Why do I need to provide my own API keys?

What is the Open Knowledge Format (OKF)?

What voice providers does TalkyTalky support?

Can I preview voices before choosing one for my partner?

What is WaveNet?

Can I bring my own Grok/xAI API key?

What models does the Tone Switcher route between?

What image generation models are available?

Is the Gallery available to all users?

What tools can my companion use, and what does each one need?

What is Deep Research, and where do I find it?