Pitfalls
Practical material on AI API relay services: how rates are calculated, how to verify a model has not been downgraded, what enterprises should check before buying, and the pitfalls we ran into while measuring these services ourselves.
Browse by topic
Pitfalls (14 articles)
Model retirement is not an accident: turning deprecation into a scheduled task
You open your laptop, get 404 model_not_found, and production is down. It is almost always an upstream retirement, not a…
RPM, TPM and concurrency: the four real shapes of relay rate limiting, and how to find the true ceiling
Rate limiting does not always return 429. Silent queueing, silent downgrades and mid-stream cutoffs report no error at a…
Three traps in choosing a site from a leaderboard alone
The leaderboard is not the problem; treating it as the only basis is. This article covers three specific traps and how t…
Evening peak jitter: upstream, network path, or your own concurrency
When the evening peak slows you down, do not blame the relay immediately. This article gives a three-step diagnosis that…
Four ways to check a "direct official connection" or "pure resale" claim
"Direct official connection" is the most common marketing line and the hardest to verify. This article gives four practi…
A 30-minute emergency checklist after an API key leak
Once a key leaks, the charges run by the minute. This article gives an order of operations you can follow, plus the thre…
Five terms to check before topping up: balance, refunds and expiry, question by question
When things go wrong, the terms are your only basis, not what a support agent promised. This article gives five question…
The hidden cost of self-hosting a relay: what it costs per month in money and time
Open source and free describes the price of the software, not of the server, the domain, or the most expensive part, mai…
45.83% of endpoints failed identity verification: how to read that industry figure
The figure that 45.83% of endpoints failed identity verification is often used as proof that nearly half of relay sites…
Account throttled or banned: the common triggers for relay service risk controls
Getting limited by a relay service is rarely random. The triggers are usually observable in advance, and most of them co…
Give a relay service a health check: a 20-minute verification checklist
Before you put real money on an account, twenty minutes of structured testing tells you most of what you need. This is t…
Six signals that a relay service is about to disappear
Relay services rarely vanish without warning. These six signals can be observed beforehand, each with a way to check it…
Supporting data & tools
The articles explain the method; the pages below give you measured data and prices to check it against. Reading them together is easier — once you know how to tell whether a rate is real, you can look the official price up in the price list.