I audited 78 API requests on my own gateway. One of them used 1,086 input tokens and returned 2 tokens. Token waste is real.

I built a small API gateway for Chinese AI models (DeepSeek, etc.) and decided to audit my own token usage. Nothing fancy — just 78 test requests.

What I found:

76 successful calls → 13,103 tokens total (avg ~172 per call)

2 calls failed → invalid model name (no cost, but indicates misconfiguration)

Worst case: 1,086 input tokens → 2 output tokens. Yes, someone pasted a massive document and got a 2-token reply. This is like using a Ferrari to deliver a pizza.

If this happens at enterprise scale, the waste is astronomical. Companies are throwing money at AI APIs without knowing where it’s going.

What I’m building:
A Token Audit tool that catches:

❌ Duplicate content submissions

❌ Expensive models used for trivial queries

❌ Nighttime spikes (potential key leaks)

Would you use this for your team? Any feedback is appreciated.

P.S. This is my own test data, not customer data. Just validating the problem.