Skip to main content

AI Usage > Model Name

Written by Tom Williams

Dataset: AI Usage

Entity: AI Usage Measurement

Field ID: model_name

Type: Text

Description: The raw model string published by the provider. Kept alongside model_ref so a model Keypup has not yet mapped is never lost.

Source: App

Transformation logic: N/A

App Mapping

GitHub Copilot

the raw model value, verbatim - including auto, unknown and others

Anthropic Claude (coming soon)

the raw model string from the per-model breakdown

Cursor (coming soon)

the raw model on usage events

OpenAI Platform (coming soon)

the raw model

OpenAI Codex (coming soon)

N/A

Reporting Use Cases

The Model Name field is the provider's own model string. model_ref is the comparable version; this is the faithful one.

  • Exact model versions: Where a provider publishes a dated or pinned model string, this field keeps it while model_ref normalises it away.

  • Spotting new models: An unfamiliar value here is usually a model the provider has just released. It will show as unattributed in model_ref until it is mapped.

  • Not comparable across providers: The same model is named differently by each tool that serves it. Use model_ref for any cross-provider chart.

Did this answer your question?