Launching soon: everything you see here opens to everyone shortly. Get notified →
lippa
Launching soon: everything you see here opens to everyone shortly. Get notified →
lippa

The workspace

Everything a chat window should have become

Lippa looks familiar, a thread, a composer, attachments. What sets it apart is everything around the thread: the models, the memory, and the work that comes out of it.

{{ t.label }}

Multi-model chat

Switch models mid-conversation. Nothing gets lost.

Pick the model per message. Send the first to Claude, the next to GPT, the next to Gemini, all in one thread, all seeing the same memory. Only the perspective changes.

Regenerate with a different model: same question, same context, a second opinion in one click, with provenance on every take. Reasoning modes: Instant, Auto or Thinking, per message, on models that support it. Auto model choice: let Lippa pick the right model for each message, or choose a lane: Smartest, Quick, Writing, Coding, Research. Continue past the context limit: when a long thread fills up, one click opens a continued chat with your memory already in place. Nothing to re-paste.
Compare our pricing to the two competitors we tracked last week.
Claude Opus 4.8
Both moved since Monday. The short version: one raised entry pricing 12%, the other unbundled…
Regenerated with GPT-5.6 Sol
Same data, different read: the 12% rise mainly affects annual plans, your entry tier is untouched…
Continue the thread, or switch model Auto · GPT-5.6 Sol
conversations memory Decision Fact Open question pinned · reviewed · yours

Memory

A memory you can read, edit and trust

Lippa distils your conversations into structured memory, facts, decisions, open questions, scoped to your account, spaces and projects. Every model you use gets the same memory layer, assembled live when you start a chat.

Curate it in the Memory Explorer: confirm, edit, archive or pin an item as authoritative. Your wording is protected, the pipeline never silently overwrites it.

The full memory story →

Launchpad & intent router

A homepage that knows what Lippa can do

The Launchpad is built around one composer: just say it. The intent router reads your goal and proposes the right capability, a document review, a consensus debate, a living dashboard, with its recommendation flagged and its confidence stated.

Power users get an express lane straight to named tools. New users learn by watching. Lippa names what it did afterwards.

How routing works →

What would you like to work on?

Keep an eye on EU AI Act guidance and tell me what changes
Monitored dossier · recommended Research brief Just answer

Nothing runs without your confirmation.

Models, combined

Put three minds on it, not one

Consensus for hard decisions, Roundtable for fast multi-angle answers, Document Reviewer for drafts that matter. Each runs several frontier models against each other, with your project memory in scope.

Consensus

Three frontier models argue it out over rounds, concede where the evidence is strong, and converge, with disagreements made explicit.

Roundtable

Reasoner, domain expert and contrarian answer at once; a fourth model writes the research brief. One toggle, one pass, many angles.

Document Reviewer

Three independent reviewers, merged into prioritised feedback and a revised draft, sharper inside a project, where your decisions are known.

Deep dive: models working together →

Spaces & projects

One more layer than a list of chats

Projects are pinned workstreams, conversations, documents and memory that belong together. Spaces group projects into areas of your life or business, and are where you invite collaborators.

A Space remembers everything you told it, and nothing you didn't. Memory stays scoped to where the work lives.

Space · Company

Project · Q3 launch Project · Pricing review Project · Hiring

Space · Personal

Project · House renovation Project · Studies

Each project carries its own memory, documents and conversations.

From chat to artifacts

The output is work, not just words

Anything substantial comes back as a versioned artifact you can reopen, edit, share and save into project knowledge.

Code Cloud

A coding agent that builds and runs real apps in your browser, and fixes its own mistakes.

Explore →

Video & images

Short clips and images from a prompt, with the exact cost shown before anything runs.

Explore →

Website Creator

Describe it, Lippa builds the pages, published and hosted on a Lippa subdomain, custom domains on the way.

Artifacts & sharing

Docs, diagrams, pages and code, versioned, editable, and shareable as unlisted links that never expose your memory.

Import your history

The browser extension brings the work you already did in ChatGPT, Claude and Gemini into Lippa's memory, so day one isn't a blank slate.

Costs, before not after

Frontier messages and actions show their credit estimate up front. Standard messages show nothing, because they cost nothing on paid plans.

Yours to take

Export chats, files and memory whenever you like. Portability is a design constraint, not a compliance checkbox.

Get started free