Support Center

Get help with CacheGateway products and services. Our team is here to support your success.

Get in Touch

Frequently Asked Questions

How do I get started with CacheGateway?

Simply sign up, create an API key from your dashboard, and replace your existing provider URL with CacheGateway. No code changes required - just update your base URL to start getting monitoring, cost tracking, and semantic caching.

Which AI providers do you support?

We support OpenAI, Anthropic, and Google AI today through a drop-in proxy. Additional OpenAI-compatible providers — including Groq, Mistral, Together, and xAI (Grok) — are rolling out. Point your existing provider SDK at our base URL; there's no separate integration to manage.

How does monitoring and observability work?

Every request through CacheGateway is automatically tracked with detailed metrics including cost, latency, token usage, and AI agent behavior. View everything in real-time from your dashboard with no additional setup.

What is your data retention and privacy policy?

We maintain strict data isolation and never use your data to train models. Request logs are retained on a plan-based schedule — 7 days on Free, up to 365 days on Scale — then automatically deleted. See our Privacy Policy for complete details.

Can I use CacheGateway with my existing code?

Yes! CacheGateway is fully compatible with OpenAI SDK and other standard AI libraries. Just update your base URL - your existing code continues to work without any modifications.

Do you offer enterprise features?

Enterprise-grade features (custom SLAs, SSO/SAML, audit log integrations, dedicated account management) are not part of the current product. If you have an enterprise need, the Scale tier covers most production workloads; email founder@cachegateway.com to discuss specifics.

Support Hours

Standard Support (Free / Starter)

Monday - Friday: 9:00 AM - 6:00 PM EST

Saturday: 10:00 AM - 4:00 PM EST

Sunday: Closed

Typical response: 4-6 hours

Free: community / best-effort · Starter: 24-hour target

Priority Support (Pro / Scale)

Faster queue placement during business hours

Typical response: 1-2 hours

Pro: 24-hour target · Scale: 4-hour target

Typical times reflect recent averages and are not guaranteed. Contractual response targets by plan are defined in our Terms of Service.

Need Immediate Help?

For urgent issues affecting production systems, contact our support team.