AI assistants and search decision
Cattus AI
A single competent developer can build a useful multi-model assistant replacement (chat, model switching, doc RAG, web search) in a few dozen hours using existing APIs and open-source UI/agent projects; enterprise features and bundled multi-model billing would still justify paying Cattus for teams.
Visit website↗$16/mo
$192/yr
Read off the official pricing page.
$100one-off76 h to build
$50/mo6 h/mo upkeep
On cash alone, building overtakes the subscription at 4 seats.
Open-source builds that already do this
Every project below is open source and already does this job today. Fork one, self-host it, or take the parts you need - the build prompt further down assumes an empty file, and this is the shortcut past that. Licences differ; check the one on each card before you ship. All Cattus AI alternatives, with the arithmetic →
What a replacement has to do
- User sends prompts -> route to chosen model -> return response in chat UI (text + optional voice) -> optionally run web search or document RAG -> present images when requested.
What it still won’t have
- Dedicated hosted multi-tenant instance and enterprise SSO offering
- Vendor-negotiated access to multiple premium models bundled behind one bill
- Production polish, UX, analytics dashboard and paid support
- Any proprietary data or model optimizations Cattus may have
What remains hard
- Product polish and ongoing maintenance
First-year cost
Keep paying
Paying is—cheaper in year one.
On cash alone, building overtakes the subscription at 4 seats.
Money you would actually spend
Time you would spend
—
What you would spend
What we assumed
The verdict above measures whether you could build it. This one is only about money.
Runnable build prompt
Build a minimal multi-model AI assistant using Next.js (React) frontend, Node/Express backend, Postgres for sessions/usage, and Redis for caching; integrate OpenAI/Anthropic (or pluggable providers) for chat and embeddings, a search API (e.g., SerpAPI) for web search, and an image-generation API. In scope: (1) model-selection routing and adapter layer, (2) text chat UI with message history and optional TTS/STT, (3) document upload → embeddings → RAG search, (4) web-search aggregation, (5) simple auth + Stripe subscription enforcement, (6) minimal admin view for usage and quotas. Out of scope: enterprise dedicated instances, SSO, advanced analytics, and multi-tenant hardened deployment. Include robust error handling, retries for API calls, rate-limit protection, and unit/integration tests covering routing, embedding+RAG, and auth flows.
How we checked
How the score was reached
- Partly verdict base52
- An open-source build was found+5
- 4 cited sources+3
- Price verified on pricing page+3
- 3/3 assessment runs agreed+4
- Evidence score67
The base comes from the verdict. Everything under it is a check that either happened or did not, and each one is a fact frozen in this record rather than a judgement made at render time - so the same evidence always produces the same number.
How scoring works →Cited sources · 4
Every page the run actually retrieved.
- official productCattus - All in one AI assistant (features & pricing)
- official pricingCattus pricing (Premium 16€/month)
- open sourceopenclaw/openclaw
- open sourcearc53/DocsGPT
Integrity checks
What held up, and what did not.
