# LLM model ID directory

Find canonical model IDs and aliases with their provider, identifier type, lifecycle context, and first-party source. The directory does not infer an API identifier from a display name.

## An ID is only part of the deployment

The name on a model page, the identifier in your request and the provider serving that request are three different pieces of one setup. The directory gives you the documented identifier and the page that published it. The serving route and the snapshot you deploy stay your choice; the directory never guesses either from a display name.

- Stable ID: the provider's current documented identifier for a model, such as claude-sonnet-5. It does not establish which snapshot answers today's request or whether a partner platform uses the same string and retirement date.

- Pinned version: a dated snapshot the provider documents separately. It fixes the weights you call until that snapshot is retired, but the same snapshot can carry different retirement dates on different platforms. A pinned ID is listed only when a provider publishes one.

- Alias: a moving pointer such as gpt-5.3-chat-latest. A request resolves to whatever the provider points it at today; an alias can be marked deprecated while the model behind it is still served.

- Unresolved identity: a model with a launch post but no published API identifier, such as GPT-5.3 Codex Spark. There is no row, because there is no string to copy yet.

## What "Check this model" does

The link opens Radar with the provider and the exact ID in the link fragment and sends no referrer. Radar does not pick a serving route from the ID; the reader chooses the route there, and only routes with a published notice return a date. A row without the link has no single documented API identifier from the provider's own site or belongs to a provider the check does not cover yet.

## Related

- [AI model deprecation calendar](https://benchlm.ai/deprecations)

- [Building a custom LLM benchmark](https://benchlm.ai/blog/posts/building-custom-llm-benchmark)

Last updated: September 24, 2026

Canonical page: https://benchlm.ai/model-ids