Model alias
A model alias is a short name that can be pointed at a newer snapshot later, while a dated or pinned id stays put. OpenAI, Anthropic, and Gemini each publish an alias rule, and Amazon Bedrock's lifecycle pages do not.
Last reviewed . Publisher: SunsetDiff. A source line names the day that page was read.
The rows below are copied from this site's model records. A shutdown cell is the date stored for that provider, not a date copied from another host.
OpenAI
A short slug can be moved to a new snapshot. The changelog records those moves. chat-latest points at the latest Instant model used in ChatGPT, and that page says the underlying snapshot is regularly updated. The July 9, 2026 changelog entry says the gpt-5.6 alias routes requests to gpt-5.6-sol.
The previous snapshot keeps its dated id. Pin that id when you need the version you tested.
Sources: Chat Latest , OpenAI changelog , GPT-5.6 Sol model page , checked October 20, 2026.
Anthropic
Before the 4.6 generation, the Claude API accepts a shorter alias such as claude-sonnet-4-5. That alias points at the most recent dated snapshot for that minor version. A 4.6-generation id is the snapshot itself. Anthropic does not update the weights of an existing model ID.
The pinned-ID guarantee does not cover those convenience aliases.
Source: Claude model IDs and versioning , checked October 20, 2026.
Gemini API
The Gemini models page splits names into stable, preview, latest, and experimental. A latest name, such as gemini-flash-latest, is an alias that is hot-swapped on every new release of that variation. Breaking changes get two weeks of email before the version behind latest changes.
Firebase's model list calls gemini-2.0-flash an auto-updated alias. That page also says not to use -latest aliases, because they hot-swap and only breaking changes get the two-week email.
Sources: Gemini API models , Firebase AI Logic models , checked October 20, 2026.
Azure OpenAI, Vertex AI, and Amazon Bedrock
Azure does not use the word alias for this. A deployment has a version and a versionUpgradeOption. OnceNewDefaultVersionAvailable moves the deployment to the new default. NoAutoUpgrade does not move. That is an upgrade setting on the deployment, not a floating model name.
The Vertex model versions page read on October 22, 2026 lists a model id and a retirement date. It does not describe a latest alias. Bedrock's lifecycle pages do not describe a short alias that is later pointed at a new snapshot. InvokeModel takes a model id or an inference profile id.
Sources: Working with Azure OpenAI models , Vertex model versions , Bedrock model lifecycle , checked October 22, 2026.
Rows from this dataset
Dataset rows whose ids are short names, latest names, or the dated ids those names are compared with.
| Model | Provider | Stored status | Shutdown | Replacement |
|---|---|---|---|---|
| gpt-4o | OpenAI | active | Not published | Not published |
| gpt-4o-2024-05-13 | OpenAI | deprecated | 2026-10-23 | gpt-5.6-sol · diff |
| chatgpt-image-latest | OpenAI | deprecated | 2026-12-01 | gpt-image-2.5-sunburst · diff |
| claude-sonnet-4-6 | Anthropic | active | Not sooner than 2027-02-17 | Not published |
| gemini-2.0-flash | Google Gemini API | retired | 2026-06-01 | gemini-3.6-flash · diff |
| gpt-chat-latest | Azure OpenAI | active | Not published | Not published |
gpt-chat-latest (2026-10-07) gpt-chat-latest-2026-10-07 | Azure OpenAI | active | 2027-05-18 | Not published |