thinking
Intelligence
Designed To Evolve
pick a model, ask anything. context stays with you — it never drops.
pick a model, ask anything. context stays with you — it never drops.
Group Chat Ai is a private, multi-model chat workspace. Every conversation and every piece of memory stays in your browser — nothing is stored on any server. The models are many, the context is yours, and it learns as you go: preferences, names, projects, and history are remembered per-user and carried into every reply.
Built to be a single place to talk to the frontier of open and proprietary models — without accounts, without per-service tabs, and without your context vanishing between sessions. You pick a model, it remembers you, and you can drop into any conversation weeks later and it still knows what matters to you.
The same engine behind the chat is a standard OpenAI-compatible API. Point any tool (curl, python, your own app) at it with your own key.
base_url = "https://gc-chat.pages.dev/v1" api_key = "YOUR_KEY" curl https://gc-chat.pages.dev/v1/chat/completions \ -H "Authorization: Bearer YOUR_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "gemini-3.1-pro", "messages": [{"role":"user","content":"hello"}], "stream": true }'
Chats and the memory bank live in localStorage on your device, keyed to
this browser. The relay that reaches the models scrubs common personal data before it
leaves your device and never stores transcripts. Clear your browser data and everything
is gone.
Before you start, read and accept the terms below. You only do this once.
1. What this is. Group Chat Ai is a multi-model chat workspace that relays your prompts to third-party AI models and returns their responses.
2. Acceptable use. Don't use the service for anything illegal, to harass or harm real people, to attack real systems or infrastructure, or to generate content that violates the law where you live.
3. No warranty. AI responses can be wrong, outdated, or incomplete. Don't rely on them alone for medical, legal, financial, or safety decisions. The service is provided "as is" and "as available".
4. Capacity & changes. The service is shared. Under load, responses may be slow or queued. Features, models, and access may change or be withdrawn at any time.
5. Keys & access. API keys are for your use. Sharing keys publicly may get them revoked. Abuse may result in access being limited or removed.
6. Age. You must be 13 or older (16+ in the EU) to use this service.
1. Your chats stay on your device. Conversations and the memory bank are stored only in your browser's local storage. We never store transcripts on any server. Clear your browser data and everything is gone.
2. What leaves your device. Your prompts are relayed to third-party model providers to generate responses. Common personal data (emails, phone numbers, card/SSN numbers, keys, IPs) is redacted before it leaves.
3. No tracking. No analytics, no ad cookies, no fingerprinting. The only server-side number kept is an anonymous online-user count, which resets constantly.
4. Third parties. Model providers process your prompts under their own policies. Don't share anything you wouldn't say to a stranger.
5. Contact. Questions about this policy? Reach out via the GitHub link in the sidebar.
By clicking "I agree" you accept the Terms of Service and Privacy Policy above.