Documentation

Welcome to aiKMind

Your collective AI platform that brings together the world's leading AI models for chat, collaboration, debate, and voice roundtables. Get better answers through consensus.

1

Sign Up

Create your free account

2

Choose Mode

Chat, Collaborate, Debate, or Roundtable

3

Select Models

Pick one or more AI models

4

Start

Begin your conversation

Chat

Conversations with AI models side-by-side. Compare responses and generate images.

Collaborate

AI models build on each other's ideas and reach consensus through structured rounds.

Debate

Structured debates between AI models with a judge delivering the verdict.

Roundtable

Live voice conversations — talk to one model or run a multi-model council.

Chat Mode

The primary mode for conversations. Chat with a single model or compare responses from multiple models. Generate and edit images directly in chat.

Single Model Chat

1

Select a model

Choose from GPT-5.2, Claude, Gemini, Grok, and more from the model dropdown

2

Type your message

Enter your question or prompt. You can paste images or drag-and-drop files

3

Enable tools (optional)

Turn on Web Search, Finance, Medical, or Legal tools for enhanced responses

4

Send

Press Enter or click Send to get your response

Multi-Model Compare

1

Select multiple models

Click the model selector and check multiple models to compare side-by-side

2

Send your message

All selected models will respond simultaneously in separate cards

3

Compare responses

View responses side-by-side. Drag cards to reorder them

4

Generate images

Ask models to generate images - compare different styles and interpretations

5

Add more responses

Use the + button to get another model's perspective

Thinking Modes

Choose from 4 thinking modes to balance speed vs. reasoning depth:

Fast

Answers quickly

Standard Thinking

Balanced reasoning and speed

Extended Thinking

Enhanced reasoning for complex problems

Maximum Reasoning

Maximum reasoning for advanced tasks

Collaboration Mode

Multi-round discussions where AI models build on each other's responses to reach comprehensive conclusions.

Starting a Collaboration

1

Select Collaborate from the sidebar

Open the Collaboration mode

2

Select participating models

Choose 2-4 models for best results

3

Configure settings

Set rounds (3-5 typical), speed mode, and synthesis options

4

Enter your topic

Type your question or topic to discuss

5

Start

Press Start and watch the collaboration unfold

Best For

  • Brainstorming and ideation
  • Research and analysis
  • Complex problem solving
  • Getting diverse perspectives
  • Logo and image creation with AI consensus

Image Generation in Collaborations

When generating images or logos, collaboration mode unlocks a unique workflow:

  • 1Clarification: Models ask follow-up questions to understand your vision
  • 2Prompt Crafting: Models discuss and refine the perfect image prompt together
  • 3Generation: Each model generates its interpretation of the consensus prompt
  • 4Compare: View all generated images side-by-side in the Images tab

Debate Mode

Structured debates between two AI models on any topic. Perfect for exploring arguments and perspectives.

Starting a Debate

1

Select Debate from the sidebar

Open the Debate mode

2

Select two models

Choose one for Affirmative (FOR) and one for Opposition (AGAINST)

3

Configure settings

Set number of turns (3-5 typical) and debate format

4

Enter your topic

Type a debatable topic (e.g., 'Should AI be regulated?')

5

Start

Watch the models argue their positions

Roundtable

Live voice conversations with AI. Talk to a single model or run a multi-model council with 6 session templates.

Voice Chat — Talk to Any Model

1

Select Roundtable from the sidebar

Open the Roundtable page

2

Choose Voice Chat mode

The default mode — a one-on-one voice conversation

3

Pick a provider and voice

Choose from OpenAI, Gemini, or Grok. Select a voice that suits you.

4

Start speaking

Click the microphone orb and ask your question. The model responds in real-time voice.

5

Continue the conversation

Ask follow-up questions naturally. The full transcript is saved to history.

Roundtable Council — Multi-Model Voice Sessions

1

Switch to Roundtable mode

Toggle from 'Voice Chat' to 'Roundtable' at the top of the page

2

Choose a template

Select from Deliberation, Debate, Red Team, Quick Pulse, Expert Panel, or Custom

3

Select providers

Pick 2-3 AI providers. Each is assigned a role based on the template (e.g., Analyst, Challenger, Strategist)

4

Ask your question

Speak or type your question. The AI council begins deliberating in real-time voice.

5

Listen and review

Models speak in turn — analyzing, critiquing, and building consensus. Full transcript is saved.

Templates

Deliberation

Analyst, Challenger, and Strategist deliberate over 4 rounds: initial analysis, challenge & pressure-test, convergence, and final synthesis. Best for research and complex decisions.

Debate

Proponent and Opponent deliver opening statements and rebuttals; a Judge delivers the verdict. Best for stress-testing ideas and policy analysis.

Red Team

Present your plan — an Advocate defends it while a Devil's Advocate tries to break it over 3 rounds. Best for finding weaknesses before you commit.

Quick Pulse

Get a fast optimistic and pessimistic take on any question in just 2 rounds. Best for quick gut-checks and rapid decision framing.

Expert Panel

Pick 2-4 specialist domains (Finance, Tech, Legal, Healthcare, etc.) — each gets a dedicated AI expert. Best for cross-disciplinary analysis.

Custom

Define exactly 3 roles with custom names and descriptions to suit your specific use case.

Roundtable Features

Real-time voice with live transcription
Web search during voice sessions
Confidence levels and citations
Export as Markdown, PDF, or JSON
Interrupt any model mid-speech
Full session history with playback

Supported Providers for Voice

OpenAI

OpenAI

GPT Realtime · 13 voices

Google

Google

Gemini Live Audio · 30 voices

xAI

xAI

Grok Voice · 5 voices

Tools & Features

Enhance your conversations with powerful tools that connect AI to real-world data.

Web Search

Real-time information from the web. Perfect for current events and time-sensitive queries.

Image Generation

Create stunning images, logos, and artwork directly in your conversations. Edit and refine generated images with follow-up prompts.

Finance Tools

Real-time stock charts, market data, company financials, and stock comparisons.

Medical Tools

Research medical literature, clinical trials, FDA-approved drug information, and check drug interactions.

Legal Tools

Search legislation, court cases, federal regulations, and congressional records.

Image Generation

Create and edit images directly in your conversations across all modes:

Generate Images

Create from text descriptions

Edit Images

Refine with follow-up prompts

In Chat: Generate images side-by-side from multiple models to compare styles.In Collaborate: Models work together to craft the perfect prompt before generating.

File Attachments

Images

PNG, JPG, GIF, WebP

Documents

PDF, TXT, MD, DOC

URLs

Any web page

Synth — Audio Studio

A full audio workspace: generate speech from text, transcribe audio files with speaker detection, and translate speech across 40+ languages.

Text-to-Speech (TTS)

1

Enter or paste text

Type your text, or use an AI response. Add [emotion] tags for expressive speech.

2

Add speakers

Use up to 5 speakers with different voices — mix providers (OpenAI, Gemini, Grok) in one generation.

3

Customize style

Choose a use case (narration, dialogue, news, meditation) and style presets (cheerful, professional, warm).

4

Generate & download

Stream audio in real time. Download as WAV, MP3, or FLAC. Save versions for comparison.

Transcribe (Speech-to-Text)

1

Upload audio or record

Upload WAV, MP3, M4A, WebM, OGG, FLAC, or MP4 files — or record directly from your microphone.

2

Choose a provider

OpenAI (real-time streaming) or Gemini (emotion detection and auto-translation).

3

Enable speaker detection

Turn on diarization to identify different speakers in the audio.

4

Get results

View plain text or speaker-segmented output. Copy or download as text, SRT, or VTT.

Translate (Speech Translation)

1

Upload audio

Upload an audio file in any language, or record from your microphone.

2

Select target language

Choose from 40+ languages. OpenAI translates to English; Gemini supports any target language.

3

View results

See original and translated text side-by-side. Copy or download either.

TTS Features

48+ voices across OpenAI, Gemini, and Grok
Multi-speaker with up to 5 voices
Emotion tags and style instructions
Speed control (0.25x to 4x)
Real-time streaming playback
Version history and downloads (WAV, MP3, FLAC)

Transcription Features

Speaker diarization (who said what)
Emotion detection (Gemini)
Auto language detection
Export as text, SRT, or VTT subtitles
Live microphone recording
Supports WAV, MP3, M4A, WebM, OGG, FLAC, MP4

Translation Features

40+ target languages (Gemini)
Any language → English (OpenAI)
Side-by-side original and translated text
Auto source language detection

Vision — Image & Video Generation

Create images and videos from text prompts using leading AI generation models.

Using Vision

1

Select Vision from the sidebar

Open the dedicated image and video generation workspace

2

Enter a prompt

Describe what you want to create — be as detailed as you like

3

Choose a model

Select from Sora (OpenAI), Veo (Google), or other available generators

4

Generate

Click generate and wait for your creation. You can use reference images for guidance.

5

Share or download

Save your creation or share it with a public link

Vision Capabilities

Image generation from text prompts
Video generation (Sora, Veo)
Image editing and refinement
Reference image support
Public sharing links
Multiple generation providers

Settings

Customize your aiKMind experience.

Account

View profile, email, subscription status

Appearance

Toggle dark/light theme

Location

Set location for weather and local info

Notifications

Configure push notifications

API Keys (BYOK)

Add your own API keys securely

Custom Prompts

Create custom system prompts

Memories

Manage what AI remembers about you

Shared Chats

Manage publicly shared conversations

Icon Reference

Quick reference for icons and controls used throughout the application.

Model & Persona Selection

Model Selector

Choose AI models for your conversation

💬

Persona Selector

Choose AI personality style

Tools

Tools Menu

Open tools panel

Image Gen

Create & edit images

Web Search

Real-time web info

Finance

Stocks & market data

Medical

Research & drug info

Legal

Laws & regulations

Attachments

Attach Files

Images, PDFs, documents, any file

Fetch URL

Load content from any web page

Collaboration Controls

Rounds

Number of discussion rounds

Role Templates

Assign roles to models

GodSpeed

Fast mode (~30s)

Goldilocks

Balanced (~1-2min)

Paranoid

Thorough (~3-5min)

Debate Controls

Turns

Number of debate turns

Swap Sides

Switch PRO/CON models

Roundtable Controls

Microphone Orb

Start/stop voice input

Voice Selector

Choose voice for chat mode

Template Picker

Choose council format

Coming Soon

Exciting features we're working on. Stay tuned for updates!

Authentication

Sign in with Apple

One-tap sign-in for iOS and macOS users

Platform

iOS and Android Apps

Native mobile apps with full feature parity

On-Premise Deployment

Deploy in your own infrastructure for enterprise

Integration

OpenRouter Integration

Access 100+ AI models through unified API

Integrations

Connect to external tools and data sources

Roundtable

Audio Recording & Replay

Record and replay full Roundtable voice sessions

Chat-to-Roundtable Handoff

Seamlessly switch from text to voice mid-conversation

Attributions

aiKMind is built with amazing services and APIs.