Simple, Transparent Pricing

Pay only for what you use, or get unlimited cache hits.

💰 Save up to 60% on your OpenClaw API costs

Free Tier
Perfect for testing and small projects
$ 0 /mo
  • 10,000 cached tokens/mo
  • Basic smart routing
  • Standard latency
  • Community support
Start for Free

💰 Compare to direct API usage

See how much you save with semantic caching

Provider 1M Input Tokens 1M Output Tokens Cache Hit Rate Effective Cost
VaultCaddyCloud (Pro) $2.50 (gpt-4o) $10.00 (gpt-4o) ~60-80% ~$4.00 total
Direct OpenAI (gpt-4o) $2.50 $10.00 0% $12.50 total
Anthropic (Claude 3.5 Sonnet) $3.00 $15.00 0% $18.00 total

💰 Heavy OpenClaw users save $50-200/month!

❓ Frequently Asked Questions

How does the semantic cache work?
We use a vector database to store previous agent prompts and responses. When OpenClaw sends a similar prompt (e.g., repeating a tool call or system prompt), we return the cached response instantly without hitting the expensive LLM API.
Can I use my own API keys?
Yes! You can plug in your own OpenAI, Anthropic, or DeepSeek API keys in the dashboard. We simply act as a proxy gateway to cache and route your requests.
What is "Smart Routing"?
Smart routing automatically detects simple tasks (like formatting or basic tool calls) and routes them to cheaper models like DeepSeek-Chat or GPT-4o-mini, while reserving expensive models like GPT-4o for complex reasoning.
Can I cancel anytime?
Yes, you can cancel your Pro subscription at any time from the billing dashboard. Your access will remain active until the end of your current billing period.

Ready to cut your API bills?

Join thousands of developers saving tokens every day.