Questions, answered plainly.

What is Symposium AI?

One app for ten AI providers, using your own API keys. Chat with up to three AIs at once, compare two answers side by side, set up a debate and watch it live, generate images, video, and audio, and on the web, run agentic data analysis with a built-in second opinion. Available on web, iOS, and Android.

Which providers does it work with?

Claude, ChatGPT, Gemini, Grok, Perplexity, Mistral, Cohere, DeepSeek, Kimi, and GLM for chat. Runway for video and ElevenLabs for audio in Create Studio. Model names change monthly; the app shows the current catalog with live pricing.

Is the experience the same on all platforms?

Chat, Compare, Debate, and Create work the same on web, iOS, and Android. Analyze mode is web only. The free tier and subscription management differ by platform; see below.

How does the web app work?

Sign up, add your API keys, and you get 20 free sessions: 5 each for chat, compare, debate, and analyze. After that, a 7-day free trial, then $5.99 a month.

Is there a demo mode on web?

No. The web version needs at least one API key before you can create a chat, comparison, debate, or analysis. Your free sessions are real sessions with real models.

Are my API keys secure on web?

Your keys are AES-256 encrypted and stored server-side. Every model request goes through our proxy, so a key is never sent to or exposed in your browser. You can test a key inside the app before saving it.

How do I subscribe on web?

When your free sessions and trial are used up, you are prompted to subscribe by card. Manage or cancel the subscription in Settings under Billing.

Is there a free demo on mobile?

Yes. Demo Mode lets you explore the whole app and play recorded chats, comparisons, and debates with no account and no keys. Because of App Store policies, creating new sessions and adding API keys requires starting the free 7-day trial.

How does the 7-day trial work on mobile?

Start the trial to unlock every feature for 7 days: add keys, create sessions, use History. At the end of the trial the subscription converts to $5.99 a month unless you cancel at least 24 hours before it ends. Yearly ($49.99) and lifetime ($129.99) options are also offered.

How do I cancel on mobile?

• iOS: Settings, Apple ID, Subscriptions, Symposium AI, Cancel Subscription
• Android: Play Store, Profile, Payments and subscriptions, Subscriptions, Cancel

Cancel at least 24 hours before your renewal date. You keep access until the end of the current billing period.

What is Analyze mode?

A web-only mode where one AI, the Operator, runs a real tool-use loop against your question: it writes and runs Python, queries SQL, searches the web, and fetches data from 38 built-in connectors, then authors a single structured report. A second AI from a different provider, the Reviewer, critiques the result. You decide which critiques go back to the Operator.

Where does my data go?

Uploaded files go into a private, isolated Python sandbox for your session. It has no internet access and is deleted after 7 days of inactivity or when you delete the session. The model sees column schemas, short previews, and whatever the code prints. Model requests transit our proxy; web tools fetch public URLs server-side.

What can I export?

One report, published as PDF, PowerPoint, Word, or an interactive page, plus an Excel workbook of the data behind it with a provenance sheet. Every sourced figure carries the request, the time, and a fingerprint of the response.

Which providers can run Analyze?

Any provider with function calling: Claude, ChatGPT, Gemini, Grok, Mistral, Cohere, Kimi, or GLM. Pair any two as Operator and Reviewer.

Is it a swarm of agents?

No. One Operator and at most one Reviewer per session. The Reviewer cannot run tools or edit the report; it can only propose. Tools never run in parallel, and the only thing that bounds a run is your Stop button.

Multi-AI chat

Bring up to three AIs into one conversation so they can check each other in real time. Attach images and documents to your first message so everyone has context from the start.

Web search with citations

Claude, ChatGPT, Gemini, Grok, and Perplexity search the live web automatically on models that support it and answer with numbered sources. There is no toggle.

Compare

One prompt, two answers side by side with live streaming. Tap to continue the conversation with the one you preferred.

Debate

Pick a motion, a format (Oxford, Lincoln-Douglas, or Policy), an intensity (Friendly, Neutral, or Hostile), and two to four debaters. Watch them argue live with scoring, voting, and leaderboards. Give each debater an ElevenLabs voice to turn it into a podcast.

Create Studio

Images from OpenAI, Gemini, and Grok, up to three models at once, with photo-based refinement. Video clips through Runway. Voiceovers and sound effects through ElevenLabs. A gallery keeps every creation together.

Personalities

Eight optional personalities, from Bestie, the empathetic co-pilot, to Devlin, the respectful counter-argument artist, plus Prof. Sage, Brody, Kai, George, Scout, and the Default. They shape tone without changing the model.

History

Every chat, comparison, debate, and analysis is saved. Reopen any of them and continue where you left off; recent context is re-injected so the conversation picks up naturally.

Expert Mode

Set default models per provider and tune model-aware parameters like temperature and max tokens, globally or per session, with live pricing shown for each model.

Do I need API keys?

For Demo Mode on mobile, no. For everything else on every platform, yes. Add keys from the providers you choose, such as OpenAI, Anthropic, or Google.

Where are my keys stored?

• Mobile: on your device only, in the OS keystore. Requests go directly from your device to each provider.
• Web: AES-256 encrypted on our servers. Requests go through our proxy, so a key is never exposed in your browser.

On both platforms your keys are never visible to us in plain text.

How much will usage cost?

Each provider bills you for the tokens you use, at their published rates, with no markup from us. Prices vary a lot by model: a fast, small model costs a fraction of a frontier one. The app shows the price of every model before you send, so you choose per message. If you run frontier models all day, your API bill will look like a subscription; you will still have ten providers in one place and a second opinion on every answer.

What's the price?

$5.99 a month on every platform, with a 7-day free trial. Mobile also offers $49.99 a year and $129.99 for lifetime access. Provider usage fees are separate and billed by each provider.

What's included with Premium?

Chat, Compare, Debate, Create, and Analyze with every supported provider; web search with citations; image, video, and audio generation; attachments; Expert Mode; all eight personalities; and report exports.