Hover to scrollA single AI model gives you one reasoning path dressed up as confidence. Council convenes the world's best models — Claude, GPT, Gemini, DeepSeek and more, all via OpenRouter — to debate your question over two rounds, cross-examine each other anonymously, and hand back a chairman's verdict with a confidence-weighted vote and the disagreement laid bare.
- Per-seat pricing: Balanced 3 credits/seat + 4 synthesis (a 5-model council ≈ 19 credits), Premium frontier 8/seat + 10 — charged only for models that actually deliberate; failed models auto-refunded
- Optionally ground the debate in your own Ounie brains — every verdict cites the exact pages it used, and invented citations are dropped
- Private by default; one toggle publishes a shareable read-only verdict page. Viewing, sharing, and re-opening are free
For agents
Works with Claude, ChatGPT, and any MCP client. One server, every app: paste the gateway config and the agent can discover and run Council — and the rest of the store — through ounie_list_apps → ounie_describe_app → ounie_invoke_app.
{
"mcpServers": {
"ounie": {
"url": "https://mcp.ounie.com"
}
}
}Scoped endpoint — just this app. Point a client here when the agent should only ever see Council's tools:
{
"mcpServers": {
"council": {
"url": "https://council.ounie.com/api/mcp"
}
}
}To spend your Ounie credits on the scoped endpoint, append your app key to the URL as ?api_key=… — the header-less form clients like the Ounie AI Team need.
Credits used
| Council seat (per model, per council) | 3 credits per call |