Agent teams
Pick the cheapest provider without getting a cheaper model.
- Requests go only to providers that pass
- A signed receipt with every answer
- Failing providers skipped automatically
Is your AI provider serving the model it claims? Test it now
authenticated.si tells you if an AI model is real or fake. We check providers against the official weights, in seconds, for cents.
Backed by
How it works
You don't, until you check. SPOT checks every provider against the real model before your agent uses it.
We download the official weights and run them on our own GPUs.
We ask the provider the same secret questions and compare its answers with ours.
Your requests go only to providers that pass, each answer with a signed receipt.
Proof of Model
A snapshot shows which model a provider was serving, and when. These are the latest, straight from our GPU.
Latest snapshots
Live from our GPU
| Snapshot at (UTC) | Verdict | Score | Tokens | Cost |
|---|
Score: distance from the real model. Up to 1.00× is normal. Cost: GPU time for the check.
The problem
Search a marketplace for one open model and you find dozens of providers selling it. Some serve a cheaper copy. Today you re-prompt until the answer looks right, or pin a trusted provider and pay up to 13 times more.
of audited API endpoints failed to prove they served the model they claimed.
Read the studyof 75 endpoints on one router changed the model's chat template.
Read the studysilent model changes across 10 providers in four months.
Read the studyAccuracy
We catch compressed copies, swapped models and hidden system prompts. Across 14 open models and 67 substitutes, the real model passed every held-out run.
FP8 and 8-bit copies are often too close to the real model to catch.
Who it's for
Pick the cheapest provider without getting a cheaper model.
See how every provider serves your model.
Prove you serve the real weights.
Trust
Every verdict says Genuine or Altered, and exactly how we got there.
We're bringing trust to open AI, one check at a time. Leave your email and we'll write before launch.