---
title: Create a local evaluation flag
description: 
product: experiment
token_estimate: 376
---
# Create a local evaluation flag

> For AI agents: a documentation index is available at [/docs/llms.txt](/docs/llms.txt). Append `.md` to any page URL for markdown, or send `Accept: text/markdown`.

Create a local evaluation flag, or migrate an existing remote evaluation flag to local evaluation.

## Create a local evaluation flag or experiment

1. [Create a flag or experiment](https://amplitude.com/docs/feature-experiment/workflow/feature-flag-rollouts#create-a-new-flag), but select **Local** for the **Evaluation Mode** setting.
2. In your new experiment, set your server-side deployment, configure the allocation (for example, target 100% of users), and activate your flag.

> **Note:**
>
> Your local evaluation experiment is ready to use with a [local evaluation SDK](https://amplitude.com/docs/feature-experiment/local-evaluation).

## Migrate an existing remote flag to local evaluation

To transition an existing remote flag to local evaluation mode, change the flag's settings and update the bucketing key.

> **Warning:**
>
> Migrating a flag from remote to local evaluation mode changes the bucketing behavior. Deactivate the flag before you migrate it. Confirm that all targeting rules follow the limitations of local evaluation.

1. In the _Advanced Settings_ section of the _Configure_ tab, change the flag's **Evaluation Mode** to **Local**.
2. Update the **bucketed by** setting from _Amplitude ID_ to _Device ID_ in each targeted user segment and in the _All Other Users_ section.