Works with
OpenAI
Anthropic
Gemini
Mistral, Groq, and local models — coming soon.
Early access is invite-only. Tell us a bit about you and we’ll send an access code to your inbox within 1–2 business days.
Council brings your models into one interface. Chain prompts into scheduled workflows. Get alerted when they drift. All on your machine.
Works with
OpenAI
Anthropic
Gemini
Mistral, Groq, and local models — coming soon.
Ask once. Council Mode sends your prompt to every model side-by-side. Pick the best answer — or take the best parts from each.
Early access feedback
“Saved me 30% of my time on weekly research. I used to rebuild context across three tabs every Monday — now it’s one app.”
Alpha user
“I finally trust what I’m shipping. I can see when my prompts drift and fix them before anyone else notices.”
Alpha user
“I can run a full competitive-intel report in minutes. That used to be a half-day of copy-paste.”
Alpha user
Run multi-model literature reviews. Keep every conversation. Search across all of them. Build repeatable research workflows.
Chain prompts into workflows. Test across models. Detect when outputs drift. Own your AI infrastructure.
Stop copy-pasting between tabs. One window for ChatGPT, Claude, and Gemini. All your research in one place. Switch models when you hit a limit.
Install the extension. Council pulls conversations from ChatGPT, Claude, and Gemini automatically. Search across all of them. Keep them forever.
String prompts together into repeatable workflows. SWOT analysis, market research, competitive intel — schedule them to run every Monday at 9am.
The workbench lets you test any prompt against any model. Version your prompts. Run them on a schedule. See exactly when and how outputs change.
Under the hood
Nine capability areas. One app. All on your machine.
Sync from ChatGPT, Claude, and Gemini. Search, favorite, organize across providers.
Bundle conversations, files, and URLs into contextual workspaces with shared instructions.
A searchable library of saved prompts with versions, tags, folders, and usage stats.
Multi-step prompt workflows. Each step’s output flows into the next via template variables.
Run any prompt or chain on cron. With drift detection on every result.
Chain runs saved as structured documents. Roll them up into templated master reports.
Code, HTML, SVG, and Markdown auto-extracted from responses. Versioned, downloadable.
Generate with DALL-E and friends, or upload for vision models. Browsable gallery.
Custom AI personas with system instructions. Or sync your ChatGPT custom GPTs directly.
| Without Council | With Council |
|---|---|
| 3 browser tabs, 3 chat histories, 3 subscriptions | One interface, unified search, your machine |
| Copy-paste between models to compare answers | Council Mode — same question, 3 models, side-by-side |
| Re-run the same research prompt every Monday manually | Schedule it. Get alerted when the output changes. |
| Your prompts live in notes or scattered docs | Versioned, testable, chainable, shareable |
| Conversations disappear when you cancel a provider | Yours forever. Export anytime. |
| Provider goes down? You stop working. | Switch models mid-conversation. Keep moving. |
| Hit the free tier limit on one model, start over on another | Switch models mid-conversation. Keep your thread. |
| Third-party chat tools take a cut on every token | Your API keys. Your bill. No middleman markup. |
Your conversations. Your models.
Your machine. Not theirs.
Currently in alpha. Bring your own API keys. Self-hosted via Docker.
Council is the interface. You own the API keys. Pay providers directly — no middleman markup.
Click request early access at the top of the page and tell us who you are. We’ll send an access code to your inbox within 1–2 business days.
Yes. Council connects to providers using your existing API keys. You pay providers directly — no markup, no middleman. Your keys stay on your machine.
Third-party tools look cheaper because they charge a flat monthly fee — but they mark up every token on top of what providers charge them. With Council, you pay OpenAI, Anthropic, or Google directly at their published rates. No middleman markup. For most power users, pay-as-you-go API usage costs 30-70% less than stacking three monthly chatbot subscriptions — and you only pay for what you actually use.
On your machine. Council syncs conversations locally. Nothing is sent to our servers. You can export everything at any time.
Yes. Council ships as a Docker container. Run it on your own infrastructure — your laptop, your server, your cloud. Full control.
Your conversations stay. They were always yours. Export them and take them anywhere.
No. Keep your existing subscriptions. Council works alongside them — syncing your conversations into one place. Cancel a provider only when you're ready.
Today: OpenAI (GPT-4o, GPT-5), Anthropic (Claude Opus, Sonnet, Haiku), and Google (Gemini 2.5 Pro). Coming soon: Mistral, Groq, DeepSeek, and local models via Ollama. Bring your own API keys for each provider you want to use.
Join the alpha and help shape the future of multi-LLM workflows.
Works with ChatGPT, Claude, and Gemini. Bring your own API keys.