Three AI models.
One conversation.
Chat with Claude, GPT, and Gemini. Compare responses, deliberate together, or assign specialist roles. 20 free messages a day.
Everything that works, right now.
What you get from day one. No waitlist, no early-access tier, no gated beta — every capability below is shipped and production-ready.
Multi-provider chat
Claude (including Opus 4.6 on Plus), GPT-4o, Gemini, o3. All the major frontier and standard models in a single conversation.
Generous daily credits
20 messages every day on Free, 150/day on Plus. Top-ups (£5 / 100 credits) work on either tier when you need more before midnight UTC.
Real Stripe billing
Subscribe, cancel, top up, manage payment methods — all from one settings page. Stripe Customer Portal handles the rest.
Privacy and control
Full conversation history downloadable as JSON anytime. Permanent account deletion that actually deletes — GDPR Article 17 compliant.
Three modes for three kinds of question.
Quick lookup? Compare perspectives? Get a fully reasoned answer? Pick the mode that fits the question.
Single chat
One model, fast. Pick any provider per message. Switch mid-conversation. The default for most questions.
Group chat
Multiple models respond to the same message sequentially. Each sees the previous response. Compare perspectives from Claude, GPT, and Gemini on the same question.
Group Thinking
Models deliberate among themselves before you see anything. One leads, others critique and find flaws, lead synthesises. One refined answer backed by collective AI reasoning. Optional transcript view.
Cowork mode — AI with specialist roles
Assign specialist roles to each model slot. One slot reasons, another writes code, another shapes the presentation. Each slot uses the best model for its job — drag-and-drop to assign, grouped by provider, save reusable configurations per project.
Drag-and-drop slots
Assign models to roles by dragging. Models grouped by provider so they're easy to find.
Same model, many roles
Use one model in multiple slots with different role contexts — a Sonnet that reasons differs from a Sonnet that codes.
Save as preset
Reusable workspace configurations per project. Switch between research mode, coding mode, design mode in one click.
Everything you need. Nothing you don't.
The deeper feature inventory. Built for the people who actually use AI day-to-day.
Three providers, one interface
Anthropic, OpenAI, Google. Pick the right model per message — no tab-switching, no multiple subscriptions.
Frontier when you need it
Claude Sonnet 4.6, GPT-4o, Gemini 2.5 Pro, o3, and Claude Opus 4.6 on Plus. 15 frontier messages a day.
No subscription to start
20 free messages every day. Top up £5 for 100 extra credits when you need more — works on any plan.
Your data, your control
Every conversation downloadable as JSON. Delete your account permanently any time, no support email required.
GDPR data export
One-click JSON export of every conversation, message, billing record, and credit ledger entry. Article 20 portability built in.
Permanent account deletion
Delete actually deletes — your auth row, conversations, and PII go. Anonymised audit trail preserved per Article 17 § 3(d).
Disposable email blocking
Signup refuses 600+ disposable email services automatically. Real users get through; abuse gets stopped at the gate.
Settings that work
Display name, password reset, billing portal, data export, account deletion — all in one page. No support tickets needed.
Simple, transparent pricing
No hidden tiers, no per-seat math. Pay for what you use.
Free
£0forever
- 20 messages/day
- Standard models: Haiku 4.5, GPT-4o mini, Gemini 2.5 Flash-Lite
- Top-ups available (£5 / 100 credits)
- Full conversation history, downloadable as JSON
- No credit card required
Plus
£20per month
- 150 messages/day
- All standard models
- Frontier (15/day): Sonnet 4.6, GPT-4o, Gemini 2.5 Pro, o3, Claude Opus 4.6
- Top-ups available (£5 / 100 credits)
- Group Thinking + Cowork mode when they ship
Pro
£200per month
- Higher daily limits (~500 messages/day)
- All Plus features
- Claude Opus 4.7 access
- Team workspaces
- API access for developers
What's next.
The sequence ahead. No specific dates — we ship when it's ready.
Group chat + deliberation
Group chat mode (multiple models, sequential responses, compare perspectives). Group Thinking on Plus (internal deliberation, one synthesised answer).
Cowork mode
Slot-based agent workspace. Specialist roles per slot — reasoning, coding, design. Drag-and-drop assignment, reusable presets.
Pro tier + API
£200/month Pro tier. Higher daily limits, Opus 4.7, team workspaces, API access for developers building on top of the multi-model routing layer.