Enterprise-grade AI API gateway with subscription plans and active-active clusters.
Deep refactor of one-api designed for enterprise API management rather than proxies.
one-api-pro is an enterprise-grade AI API gateway built with Go and Vue 3. It provides subscription billing, token management, and decentralized active-active clustering for organizations managing multiple AI models.
The platform unifies over 30 model providers behind an OpenAI-compatible interface, offering fine-grained token controls and precise cost accounting. It includes built-in support for WeChat Pay and Alipay for real-time quota top-ups and subscriptions.
It suits enterprise engineering teams managing internal AI usage across multiple teams or departments. It is not suitable for public AI relay or forwarding stations.
Nodes run independent MySQL and Redis instances, syncing via application-layer events without a shared database.
Provides token or request-based billing, period rate limits, and upgrade pricing based on remaining days.
Enforces per-model allowlists, sub-net IP restrictions, and quota caps down to individual models.
Supports WeChat Pay Native and Alipay Face-to-Face out of the box for automated user top-ups and orders.
The software is open source with no stated licensing fees in the repository metadata. The only costs are the servers, databases, and payment gateway transaction fees required to run it yourself.
767 open remote AI-training roles, updated November 14, 2026; most pay $20–$175 an hour. Zekai lists them free; the application goes to the employer’s platform.
No, the developers explicitly state that this application targets enterprise AI gateway use cases and is not built for public relay stations.
Each node runs its own MySQL and Redis instances while application-layer events keep data synchronized across the cluster.
See Zekai first in Google