Astralane / Developers
Your first request
Create a workspace, fund a development wallet, issue a key, and call the local demo provider.
1. Start the local stack
Follow README.md: generate .env, start PostgreSQL and Redis, install dependencies, migrate, seed, and run pnpm dev. The web app runs on port 3000, the gateway on 4000, and the local demo provider on 4100.
node infra/scripts/setup.mjs
docker compose up -d
pnpm install
pnpm db:migrate
pnpm db:seed
pnpm dev2. Create an account and fund the wallet
Sign up, open Billing, and select Add $10 test funds. The grant is available once per development workspace only. Production has no development grant.
3. Create an API key
Open API keys, create a key with the inference scope, and copy its plaintext value immediately. It cannot be retrieved later. Keep it in an environment variable or secret manager.
4. Make a real local HTTP request
The development model is a clearly labeled deterministic mock. It exercises routing, metering, and settlement but does not call a real AI provider.
curl 'https://api.astralane.ai/v1/chat/completions' \
-H "Authorization: Bearer $ASTRALANE_API_KEY" \
-H 'Content-Type: application/json' \
-d '{"model":"demo/nexlane-small","messages":[{"role":"user","content":"Hello Astralane"}],"max_tokens":128,"stream":false}'5. Verify the outcome
Open Usage and Billing. The completed request should include tokens, exact cost, synthetic demo savings, and timing. The ledger should contain reservation, release, and charge entries.
Native core and mock HTTP checks have run. Full dependency install, PostgreSQL/Redis integration, production builds, and browser flows still require local verification. See the repository verification report before deployment.