AI assistants and search decision
Bae by ppl.studio
A capable developer can build a usable AI companion (chat, memory, basic onboarding, payments) within a multi-week effort using existing OSS assistants, but reproducing Bae's polished archetypes, proprietary multi-layer memory system, safety/age-verification and photo aesthetics would be expensive and time-consuming.
Visit website↗$8.23/mo
$99/yr
Read off the official pricing page.
$100one-off90 h to build
$50/mo6 h/mo upkeep
On cash alone, building overtakes the subscription at 8 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 Bae by ppl.studio alternatives, with the arithmetic →
What a replacement has to do
- Create companion (quiz/archetype) → customize face/name/voice → chat session with persistent memory → occasional selfies and weekly digests
What it still won’t have
- Polished, hand-written archetype library and voice tuning
- The product's six-layer proprietary memory pipeline (constitution, canon, retrieval, reflection, drift detection)
- The exact selfie/scene-photo aesthetic and consistent person imagery
- Age-verified / consent-first Spicy mode & built safety workflow
- Product polish, UX refinements, and brand-trust
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 8 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 AI companion web app using Next.js + React frontend, FastAPI backend, Postgres for persistent memory, S3-compatible object storage for selfies, and OpenAI-compatible LLM API. Core features in scope: 1) 2-minute onboarding quiz + archetype chooser, 2) companion creator (name, voice register, face placeholder), 3) chat UI wired to LLM with per-message memory writes and a memory retrieval endpoint that returns pinned facts and recent context, 4) two-tier memory retention (7-day rolling for free, persistent for Pro), 5) Stripe subscription gating for Pro features, 6) selfie upload and storage, 7) scheduled worker to advance simple relationship stages. Out of scope: advanced six-layer memory pipeline, age-verification, Spicy mode, bespoke consistent-person photo synthesis. Include input validation, error handling, unit and integration tests for major APIs, and Docker deployment manifests for hosting on a single VPS; document setup and environment variables.
How we checked
How the score was reached
- Partly verdict base52
- An open-source build was found+5
- 5 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 · 5
Every page the run actually retrieved.
- official productBae — home
- official pricingBae — pricing
- official docsBae — features
- open sourceleon-ai/leon
- open sourceagentscope-ai/QwenPaw
Integrity checks
What held up, and what did not.
