Gemini + RevenueCat + Supabase in a React Native boilerplate
Does AI Mobile Launcher support this combo?
Yes. Gemini is one of two cloud LLM providers in the AI Pro tier, alongside OpenAI. The REST client lives at src/features/gen-ui/api/gemini.api.ts. RevenueCat is wired with a paywall, free trial, and restore flow. Supabase handles auth + Postgres + RLS. The combo also powers the multimodal analyser (audio / image / video) because Gemini accepts mixed inputs in a single call.
The stack
- React Native 0.83.6 + Expo SDK 55.0.17
- Gemini: REST client at
src/features/gen-ui/api/gemini.api.ts. Defaultgemini-2.0-flash, swap togemini-2.0-profor tougher reasoning - RevenueCat:
react-native-purchasespre-wired - Supabase:
@supabase/supabase-js+ RLS + pgvector available - Multimodal: same Gemini key powers audio / image / video analysis
Setup in five steps
1. Clone and install
git clone <ai-pro-tier-repo>
cd ai-mobile-launcher
pnpm install2. Gemini key
# .env
GOOGLE_GEMINI_API_KEY=AIza...
GEMINI_MODEL=gemini-2.0-flashGet the key from aistudio.google.com. Free tier covers prototyping; switch to paid for production rate limits.
3. Supabase
# .env
EXPO_PUBLIC_SUPABASE_URL=https://<project>.supabase.co
EXPO_PUBLIC_SUPABASE_ANON_KEY=<anon-key>4. RevenueCat
# .env
EXPO_PUBLIC_REVENUECAT_IOS_KEY=appl_...
EXPO_PUBLIC_REVENUECAT_ANDROID_KEY=goog_...5. Run on a device
eas build --profile development --platform iosWhy this combo works
Gemini is the cheapest cloud LLM that punches at GPT-4o quality for most mobile workloads. The 1M token context window is overkill for chat but ideal for document Q&A and the kind of long-history personalization onboarding flows need.
Multimodal is the differentiator on mobile. Most mobile features that need AI are camera-based: calorie tracking, plant ID, document scan. Gemini takes an image plus a prompt in one call. With OpenAI you wire Vision separately; with Gemini it is the same endpoint.
RevenueCat handles store-receipt validation so you do not. Supabase pairs with Gemini because pgvector ships in every project, which is useful when you start building RAG features against user content.
What it costs at scale
| Line item | 1K MAU | 100K MAU |
|---|---|---|
| Gemini 2.0 Flash (10 chats, 500 tokens) | ~$1 | ~$100 |
| RevenueCat (free up to $2.5K MTR) | $0 | ~$200 |
| Supabase Pro | $25 | ~$125 |
| Total infra (excl. store fees) | ~$26 | ~$425 |
Gemini Flash is the cheapest cloud LLM on this comparison page at parity quality. If your feature is multimodal, it is the default choice on cost alone.
What this combo does NOT cover
- Offline / on-device: use llama.rn for that
- Image generation: Gemini analyses images, does not generate them
- Web payments: RevenueCat targets iOS / Android stores; for web Stripe see OpenAI + Stripe combo
- Anthropic / Claude: see that combo page; you wire it manually
Get this combo
Ships in the AI Pro tier ($199).
See AI Pro tier ($199)