Published · updated

About Ofox.ai

Ofox.ai is a unified AI API gateway: 100+ models behind one key, on the OpenAI, Anthropic and Gemini protocols, pay-as-you-go with a balance that never expires.

Ofox

Ofox.ai is a unified AI API gateway. One API key and one balance reach 100+ models across the GPT, Claude, Gemini, DeepSeek, Kimi, Qwen, GLM, Grok, MiniMax and Doubao families, plus image and video models.

The gateway speaks three protocols: OpenAI, Anthropic and Gemini. Point your existing base URL at the matching Ofox endpoint, swap in an Ofox key, and the rest of your code keeps working with no SDK change. The OpenAI-compatible endpoint is https://api.ofox.io/v1.

What Do You Get with Ofox.ai?

  • One integration - three protocols behind one key, and switching models is a one-string change
  • Pay as you go - no subscription, no monthly fee, and a balance that does not expire
  • 0% platform fee - billed at each model vendor's official rates
  • Automatic fallback - when the primary model errors, the request retries down your fallback list, with provider-level routing available
  • Usage and budgets - set daily, weekly or monthly budgets per key or per user in the console, with automatic limiting and cost alerts
  • Global delivery - edge nodes across multiple regions

How Do You Get Started?

  1. Sign up in the console and create an API key
  2. Repoint the base URL in your code to the matching Ofox endpoint
  3. Run your first request with the quickstart

Where Do You Find Pricing and Docs?

  • Model catalog - live pricing and capabilities per model
  • Developer docs - authentication, streaming, function calling, prompt caching, error handling, fallback
  • For teams - usage, budgets and support response times; compliance and deployment questions go through sales
  • Privacy policy - how request content is handled and forwarded

Frequently Asked Questions

How is Ofox.ai different from calling each vendor API directly?
The integration is the same; the accounts and the bill collapse into one. Point your base URL at the matching Ofox endpoint, swap in an Ofox key, and your SDK and the rest of your code keep working. Switching models is a one-string change. Billing follows each model vendor’s official rates with a 0% platform fee.
Is there a subscription or a monthly fee?
No. The billing page in the developer docs states there are no subscriptions, no plans, no monthly fees and no minimum spend. You top up, start using it immediately, and are charged on actual usage settled in real time.
Does my balance expire?
No. The billing page in the developer docs states that the balance never expires, so you can top up as needed without a use-it-or-lose-it window.
What happens when an upstream model provider goes down?
You can configure automatic fallback: when the primary model returns an error, the request retries down the fallback list you define, and provider-level routing lets you pin traffic to a specific upstream. Both are documented under Fallback and Provider Routing in the developer docs.
How do teams control usage and cost?
The console takes daily, weekly or monthly budgets per API key or per user, with automatic rate limiting once a budget is hit. Usage breaks down by model, provider and key, and four alert types — cost, error rate, latency and fallback frequency — can be delivered by email or webhook to Slack, Lark or DingTalk.
← All posts