Account pool · live model registry · OpenAI-compatible router
utim.dev account pool with live free-model routing · auto-rotation · health probes
| Status | Plan | Slot / Credits | Bonus | Key | Created |
|---|
Disabled accounts are skipped by the router. Accounts auto-disable after 5 consecutive upstream failures and re-enable on toggle.
/v1/models
UTIM enforces a global daily request cap per free model (shared across all accounts) — a 429 "free-models-per-day" model resets at 00:00 UTC. The background probe re-checks every 30 minutes and runtime 429s instantly hide a model.
Requests round-robin across healthy pool accounts; exhausted accounts are skipped automatically.
—•••••loading…curl -s https://…/v1/chat/completions \
-H "Authorization: Bearer <key>" -H "Content-Type: application/json" \
-d '{"model":"minimax/minimax-m3:free","messages":[{"role":"user","content":"hi"}],"stream":true}'
# 1) Get an admin token (12h TTL)
curl -s -X POST https://…/api/admin/login \
-H "Content-Type: application/json" \
-d '{"key":"<admin-key>"}'
# → {"ok":true,"token":"***"}
# 2) Use the token on any /api/* call
curl -s https://…/api/accounts -H "X-Admin-Token: <token>"
curl -s https://…/api/models/live -H "X-Admin-Token: <token>"
curl -s -X POST https://…/api/models/refresh -H "X-Admin-Token: <token>"
curl -s -X POST https://…/api/accounts/sync -H "X-Admin-Token: <token>"
curl -s -X POST https://…/api/accounts/<id>/toggle -H "X-Admin-Token: <token>"
curl -s -X POST https://…/api/accounts/delete -H "X-Admin-Token: <token>" \
-H "Content-Type: application/json" -d '{"ids":["<id>"]}'
# OpenAI traffic uses the pool key instead:
# POST /v1/chat/completions · GET /v1/models
# Authorization: Bearer <pool-key>