Frequently Asked Questions

ParliAI is an AI council platform. You ask a question and get answers from multiple AI providers, debated, scored, and compared in one place. Instead of checking each model separately, you see all perspectives side by side with full reasoning trails.

Models from Anthropic (Claude), OpenAI (GPT), Google (Gemini), and xAI (Grok). New models are added regularly as providers release them.

If your preferred provider isn't listed, let us know at support@parliai.com. User requests directly influence which providers get added next.

Credits map to LLM compute cost. Different models cost different amounts: a small, fast model uses fewer credits than a large frontier one. Express councils cost less than premium ones. Your plan determines how many credits you get each billing cycle.

Three modes, depending on how much deliberation you need:

• Consensus: multi-round debate. Agents propose, voters vote, weaker proposals get dropped, and the process repeats until the council converges on a winner.
• Curator Pick: single round. Agents propose, the curator scores and picks the best one. Faster and cheaper than consensus.
• One-on-One: direct chat with a single model. No council, no debate, just a straightforward conversation.

Yes. Pick models from any supported provider and assign them roles: agents propose solutions, voters vote on them, and the referee (curator) curates the process. You can set custom system prompts for each member and save the whole configuration as a reusable preset.

If you already cross-check important decisions across multiple models, ParliAI does that for you in one place with a single subscription. Instead of subscribing to multiple AI services and copy-pasting your question between tabs, you get structured debate, voting, and transparent reasoning trails, all compared side by side.

Your prompts and responses are stored so you can view your conversation history. We don't train models on your data. Prompts are sent only to the providers in your council configuration, and you control which providers are used. For full details, see our Privacy Notice.

There is an API, but it's currently unstable and undocumented, so we don't officially support it yet. Authentication is handled through standard JWT auth via magic email links or Google OAuth — we don't support API tokens at this time.

If you're interested in API access, we'd love to hear from you at support@parliai.com. Your feedback will help us prioritize building it out properly.

There's no free tier right now. This is a small side project funded out of pocket, and every decision costs real money in LLM compute. The Mini plan starts at $5/month for light usage if you want to try things out without a big commitment.