◈ PATHOTELS
Login
AI infrastructure · API-first platform

Build with AI.
Ship with confidence.

One developer console for API credentials, model discovery and inference testing. Connect your applications to Pathotels as services become available.

Developer overview

API status

Checking…

Session

Not signed in

GPU

Not verified live

Sign in, enter your authorized organization ID under API keys, then load your dashboard. Real inference remains disabled until approved GPU deployment.

Account

New registration requires verified email and backend mail delivery; existing admin accounts may sign in directly.

Register new account
Forgot password

Organization dashboard

Shows only data authorized for your organization and membership. Platform-wide registry is private.

Wallet: sign in and load your organization dashboard. Billing acceptance pending.

Recent usage

No organization usage loaded.
Sign in and enter an authorized organization ID under API keys.

API key management

Sign in to load your own API keys. Secrets are never listed.

Secret is returned once. Copy it immediately; do not share it.

Models & provider registry

Customer model availability is read from your organization-scoped dashboard. An empty list means no customer model entitlement registry is configured. Platform registry remains private.

Only organization-authorized models are displayed. An empty list does not mean a model is deployed.
No model entitlements have been verified.

Developer API documentation

Live AI inference is not available while GPU deployment is locked. API keys and account operations must be tested only against authorized organizations.

Authentication

Create an API key in the console. Send the key in the Authorization: Bearer <YOUR_API_KEY> header. Never embed your key in public frontend code or publish it in a repository.

Chat completions

Endpoint: POST /api/v1/chat/completions. The requested model must be enabled and permitted for your organization. While GPU is locked, inference requests are rejected without executing a model.

curl -X POST 'https://YOUR_PATHOTELS_DOMAIN/api/v1/chat/completions' \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -H 'Content-Type: application/json' \
  -H 'Idempotency-Key: unique-request-id-12345' \
  -d '{"model":"YOUR_ENABLED_MODEL_KEY","messages":[{"role":"user","content":"Hello"}]}'

Replace placeholders with your own approved domain, key, and enabled model. The idempotency key should be unique per logical request; reuse only for an exact retry. A 401 indicates missing/invalid credentials, 403 denied access or disabled inference, 409 conflicting request identity, and 429 rate limiting. Do not automatically retry ambiguous 503 outcomes with a different idempotency key.

Account and dashboard

Sign in through the Account tab. Your dashboard and key operations are scoped to an authorized organization; never use another organization's ID. Available models are shown only when the backend has verified customer entitlements.

Billing safety

Wallet balances and usage are informational. Commercial inference and automatic reconciliation remain unavailable until separate security and billing acceptance is complete.

API playground

Uses /api/v1/chat/completions. The gateway may be disabled until a remote GPU endpoint is connected. Usage/cost verification is still required before paid launch.

Super Admin

Administrative access is enforced by the existing backend. This UI does not grant or bypass platform permissions.

GPU administration requires a separate active platform Super Admin authorization. This page never grants that permission.

Technical response (advanced)
Ready.