AI Governance
Logo
Model Registry Troubleshooting & FAQ

Model Registry Troubleshooting & FAQ

Logo

3 mins READ

Introduction

Model records behave like agent records with two differences that account for most of the confusion: a model is measured in spans rather than traces, and it has nothing to instrument.

A model shows no spans

The empty state says spans appear once the model starts processing requests. Several different situations produce it, and they need different responses:

Situation

How to tell

Do

The platform does not collect usage for it

It is not an OpenAI model

Nothing to fix on your side. Usage collection for models exists for OpenAI today

Its completions are not stored

An OpenAI model whose completions your application does not ask OpenAI to store

Store the completions with OpenAI so the platform can collect them

The connection is not healthy

The connection the model was imported through is inactive or failing

Fix the connection

Nothing calls it

No application uses this model

Nothing — it is inventory, not a fault

Note: A model's spans come from its provider, not its callers. The platform collects a model's usage from the provider, so instrumenting an agent does not fill a model's spans. Check the model's platform, whether its completions are stored, and its connection before you look at agents.

A model is missing from the import list

The list is the connection's answer to "what can you serve", asked at that moment. If a model is not in it, the connection cannot see it — its credentials are scoped more narrowly than you expect, or the platform does not expose that model to this account.

"This connection returned no importable models" means the connection is working and has nothing to offer. That is a different problem from a connection that fails.

Model cost looks wrong

  1. Are you looking for it in the agent total? — Spans that belong to a model asset are counted only in the Model view, not in the Agent view — see Agent View vs Model View.

  2. Is there a rate for this model? — Usage is observed either way; cost needs a rate. A model with none shows activity and no spend.

  3. Are token counts reported or derived? — Derived counts are an approximation, and accuracy varies by integration rather than by model.

  4. Are you comparing to an invoice? — They answer different questions. See How Cost & Tokens Are Calculated.

FAQs

Why does a model have five tabs when an agent has seven?

No Setup (nothing to instrument), no Monitoring, no Outcomes (an outcome is credited to the agent that accomplished it).

Why Inactive rather than Setup Pending?

Because there is no setup. A model you import shows Inactive until the platform confirms data for it. Check Last span and Activity to see whether spans are arriving.

Can a policy cover a model?

Yes — the Applies to tab covers agents and models, and the asset picker filters by type.

Should I import every model the connection offers?

No. Import what is actually called. An inventory of unused models is noise in every downstream report.

Why does the same model appear under two connections?

Two credentials both reach it. They are genuinely two paths, and importing both is sometimes correct.

Who owns an imported model?

Import sets the owner to the person who ran it. There is no owner edit on the model record today, so import from the account of the person who should own the models.

Why is activity in spans rather than traces?

A model has calls, not runs. One model can appear in spans belonging to many different agents' traces.