> For the complete documentation index, see [llms.txt](https://docs.arcee.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.arcee.ai/policies/deprecation-policy.md).

# Deprecation Policy

### Overview

Arcee models are **actively maintained and updated** to provide the latest performance and reliability improvements. Cloud-hosted versions may be **retired as part of regular updates**, while all released models remain **downloadable** on Hugging Face for independent use.

### Versioning & Updates

* Cloud models are **versioned** and may be replaced quickly when new improvements are available.
* Downloads remain stable, ensuring your workflows are not disrupted.
* Deprecated cloud versions **may stop receiving updates or support** after the notice period.

### Deprecation Criteria

A model may be retired due to:

* Release of a **new or improved version**
* **Security or reliability concerns**

### Notice & Transition

* Typical notice for standard cloud models: **up to 5 business days**
* Widely adopted models: **30 days**
* Notifications are provided via **email, API dashboard, and changelog**, with guidance for migration or upgrades.

### Commitment

We aim to **balance rapid innovation with user flexibility**, keeping cloud models current while ensuring downloadable models remain fully accessible.