 Alexandra Klepper [GitHub](https://github.com/alexandrascript) [Bluesky](https://bsky.app/profile/alexandrascript.com)  Thomas Steiner [GitHub](https://github.com/tomayac) [LinkedIn](https://www.linkedin.com/in/thomassteinerlinkedin) [Mastodon](https://toot.cafe/@tomayac) [Bluesky](https://bsky.app/profile/tomayac.com) [Homepage](https://blog.tomayac.com/)
Published: October 22, 2025
| Explainer | Web | Extensions | Chrome Status | Intent |
|---|---|---|---|---|
| [GitHub](https://github.com/explainers-by-googlers/writing-assistance-apis/) | Developer trial | Developer trial | [View](https://chromestatus.com/feature/5112320150470656) | [Intent to Experiment](https://groups.google.com/a/chromium.org/g/blink-dev/c/eLNXoxgx8CU/m/KHTF-pslBAAJ) |
The Rewriter API helps you revise and restructure text. This API and the
[Writer API](https://developer.chrome.com/docs/ai/writer-api.md.txt) are part of the
[Writing Assistance APIs proposal](https://github.com/explainers-by-googlers/writing-assistance-apis/).
These APIs can help you improve content created by users.
> [!IMPORTANT]
> **Important** : The built-in foundation model is a generative AI model. Before you build with APIs that use it, you should review the [People + AI Guidebook](https://pair.withgoogle.com/guidebook/) for best practices, methods, and examples for designing with AI.
### Use cases
Refine existing text by making it longer or shorter, or changing the tone. For
example, you could:
- Rewrite a short email so that it sounds more polite and formal.
- Suggest edits to customer reviews to help other customers understand the feedback or remove toxicity.
- Format content to meet the expectations of certain audiences.
Is your use case missing? Join the [early preview program](https://developer.chrome.com/docs/ai/join-epp.md.txt)
to share your feedback.
## Get started
[Join the Rewriter API origin trial](https://developer.chrome.com/origintrials#/view_trial/444167513249415169),
running in Chrome 137 to 148.
### Review the hardware requirements
The following requirements exist for developers and the users who operate features using these
APIs in Chrome. Other browsers may have different operating requirements.
The **Language Detector** and **Translator APIs** work in Chrome on
desktop. These APIs do not work on mobile devices.
The **Prompt API** , **Summarizer API** , **Writer API** ,
**Rewriter API** , and **Proofreader API** work in Chrome when the
following conditions are met:
- **Operating system** : Windows 10 or 11; macOS 13+ (Ventura and onwards); Linux; or ChromeOS (from Platform 16389.0.0 and onwards) on [Chromebook Plus](https://www.google.com/chromebook/chromebookplus/) devices. Chrome for Android, iOS, and ChromeOS on non-Chromebook Plus devices are not yet supported by the APIs which use foundation models.
- **Storage** : At least 22 GB of free space on the volume that contains your Chrome profile.
> [!NOTE]
> Built-in models should be significantly smaller. The exact size may vary slightly with updates.
- **GPU or CPU** : Built-in models can run with GPU or CPU.
- **GPU**: Strictly more than 4 GB of VRAM.
- **CPU**: 16 GB of RAM or more and 4 CPU cores or more.
- **Note**: The Prompt API with audio input requires a GPU.
- **Network** : Unlimited data or an unmetered connection.
> [!IMPORTANT]
> **Key term** : A [metered connection](https://support.microsoft.com/windows/metered-connections-in-windows-7b33928f-a144-b265-97b6-f2e95a87c408) is a data-limited internet connection. Wi-Fi and ethernet connections tend to be unmetered by default, while cellular connections are often metered.
> [!NOTE]
> **Note**: The network requirement is only for the initial download of the model. Subsequent use of the model does not require a network connection. No data is sent to Google or any third party when using the model.
Gemini Nano's exact size may vary as the browser updates the model. To determine the current size, visit `chrome://on-device-internals`.
> [!NOTE]
> **Note**: If the available storage space falls to less than 10 GB after the download, the model is removed from your device. The model redownloads once the requirements are met.
### Sign up for the origin trial
The Rewriter API is available in a joint origin trial with the Writer API.
To start using these APIs:
1. Go to the [Rewriter API origin trial](https://developer.chrome.com/origintrials#/view_trial/444167513249415169).
2. Click **Register** and fill out the form. In the Web origin field, provide your [origin](https://web.dev/articles/same-site-same-origin#origin) or extension ID, `chrome-extension://YOUR_EXTENSION_ID`.
3. To submit, click **Register**.
4. Copy the token provided, and add it to every participating web page on your origin or include it [in your Extension manifest](https://developer.chrome.com/docs/web-platform/origin-trials.md.txt#extensions).
5. Start using the Rewriter API.
Learn more about how to [get started with origin trials](https://developer.chrome.com/docs/web-platform/origin-trials.md.txt).
### Add support to localhost
To access the Rewriter API on localhost, use
[Chrome flags](https://developer.chrome.com/docs/web-platform/chrome-flags.md.txt):
1. Set the `chrome://flags/#rewriter-api` flag to **Enabled** or **Enabled Multilingual**.
2. Click **Relaunch** or restart Chrome.
## Use the Rewriter API
First, run feature detection to see if the browser supports these APIs.
```javascript
if ('Rewriter' in self) {
// The Rewriter API is supported.
}
```
> [!TIP]
> **Tip:** Use the [@types/dom-chromium-ai](https://www.npmjs.com/package/@types/dom-chromium-ai) npm package to get TypeScript typings for the Prompt API and other built-in AI APIs for easier access in your codebase.
The Rewriter API, and all other built-in AI APIs, are integrated in the browser.
The model is downloaded separately the first time any website uses a
built-in AI API. In practice, if a user has already interacted with a built-in
API, they have downloaded the model to their browser.
To determine if the model is ready to use, call the asynchronous
[`Rewriter.availability()`](https://developer.chrome.com/docs/ai/get-started.md.txt#model-download) function.
If the response to `availability()` was `downloadable`, listen for
download progress and inform the user, as the download may take time.
```javascript
const availability = await Rewriter.availability();
```
To trigger model download and start the rewriter, check for
[user activation](https://developer.chrome.com/docs/ai/get-started.md.txt#user-activation)
and call the `Rewriter.create()` function.
```javascript
const rewriter = await Rewriter.create({
monitor(m) {
m.addEventListener("downloadprogress", e => {
console.log(`Downloaded ${e.loaded * 100}%`);
});
}
});
```
### API functions
The `create()` function lets you configure a new rewriter object. It takes an
optional `options` object with the following parameters:
- `tone`: [Writing tone](https://owl.purdue.edu/owl/multilingual/multilingual_students/key_concepts_for_writing_in_north_american_colleges/style_genre_and_writing.html) can refer to the style, character, or attitude of the content. The value can be set to `more-formal`, `as-is` (default), or `more-casual`.
- `format`: The output formatting, with the allowed values `as-is` (default), `markdown`, and `plain-text`.
- `length`: The length of the output, with the allowed values `shorter`, `as-is` (default), and `longer`.
- `sharedContext`: When rewriting [multiple pieces of content](https://developer.chrome.com/docs/ai/rewriter-api.md.txt#multiple-tasks), a shared context can help the model create content better aligned with your expectations.
> [!NOTE]
> **Note:** Once set, the parameters can't be changed. Create a new `rewriter` object if you need to modify the parameters.
The following example demonstrates how to initiate a `rewriter` object:
```javascript
const options = {
sharedContext: 'This is an email to acquaintances about an upcoming event.',
tone: 'more-casual',
format: 'plain-text',
length: 'shorter',
};
const available = await Rewriter.availability();
let rewriter;
if (available === 'unavailable') {
// The Rewriter API isn't usable.
return;
}
if (available === 'available') {
// The Rewriter API can be used immediately .
rewriter = await Rewriter.create(options);
} else {
// The Rewriter can be used after the model is downloaded.
rewriter = await Rewriter.create(options);
rewriter.addEventListener('downloadprogress', (e) => {
console.log(e.loaded, e.total);
});
}
```
### Assign expected languages
The Rewriter API supports multiple languages. Set the expected input, output,
and context languages when creating your session.
This allows the browser to reject the request if the
browser cannot support a specific language combination.
```javascript
const rewriter = await Rewriter.create({
tone: "more-formal",
expectedInputLanguages: ["en", "ja", "es"],
expectedContextLanguages: ["en", "ja", "es"],
outputLanguage: "es",
sharedContext: "These are requests to rewrite messages to teachers in a
Spanish language program, by students who may speak Spanish, Japanese, or
English. Staff expect questions to be written in Spanish."
});
```
### Start rewriting
There are two ways to output content from the model: request-based output and
streaming.
#### Request-based output
For request-based output (or "non-streaming"), the model waits for the entire
input to be generated, process that input as a whole, and then produces
the output.
To get a request-based output, call the asynchronous `rewrite()` function. You
must include the initial text that you want to be rewritten. You can add an
optional `context` to provide the model background information, which may help
the model better meet your expectations for the output.
```javascript
// Request-based
const rewriter = await Rewriter.create({
sharedContext: "A review for the Flux Capacitor 3000 from TimeMachines Inc."
});
const result = await rewriter.rewrite(reviewEl.textContent, {
context: "Avoid any toxic language and be as constructive as possible."
});
```
#### Stream rewriting output
[Streaming](https://developer.chrome.com/docs/ai/streaming.md.txt) offers results in real-time. The output updates
continuously as the
input is added and adjusted.
To get a streaming rewriter, call the `rewriteStreaming()` function and iterate
over the available segments of text in the stream. You can add an optional
`context` to provide the model background information, which may help the model
better meet your expectations for the output.
```javascript
const rewriter = await Rewriter.create({
sharedContext: "A review for the Flux Capacitor 3000 from TimeMachines Inc."
});
const stream = rewriter.rewriteStreaming(reviewEl.textContent, {
context: "Avoid any toxic language and be as constructive as possible.",
tone: "more-casual",
});
for await (const chunk of stream) {
composeTextbox.append(chunk);
}
```
#### Share context for multiple tasks
You may want to use a `rewriter` to generate multiple pieces of content. In this
case, it's useful to add `sharedContext`. For example, you may want to help
reviewers give better feedback in comments.
```javascript
// Shared context and per writing task context
const rewriter = await Rewriter.create({
sharedContext: "This is for publishing on [popular website name], a business and employment-focused social media platform."
});
const stream = rewriter.rewriteStreaming(
"Love all this work on generative AI at Google! So much to learn and so many new things I can do!",
{
context: "The request comes from someone working at a startup providing an e-commerce CMS solution.",
tone: "more-casual",
}
);
for await (const chunk of stream) {
composeTextbox.append(chunk);
}
```
#### Reuse a rewriter
You can use the same rewriter to edit multiple pieces of content. This may be
particularly useful if adding the rewriter to a feedback or commenting tool, to
help writers offer productive and helpful feedback.
```javascript
// Reusing a rewriter
const rewriter = await Rewriter.create({
sharedContext: "A review for the Flux Capacitor 3000 from TimeMachines Inc."
});
const rewrittenReviews = await Promise.all(
Array.from(
document.querySelectorAll("#reviews > .review"),
(reviewEl) => rewriter.rewrite(reviewEl.textContent, {
context: "Avoid any toxic language and be as constructive as possible.",
tone: "more-casual",
})
),
);
```
#### Stop the rewriter
To end the rewriting process, abort the controller and destroy the `rewriter`.
```javascript
// Stop a rewriter
const controller = new AbortController();
stopButton.onclick = () => controller.abort();
const rewriter = await Rewriter.create({ signal: controller.signal });
await rewriter.rewrite(reviewEl.textContent, { signal: controller.signal });
// Destroy a rewriter
rewriter.destroy();
```
## Demo
## Permission Policy, iframes, and Web Workers
By default, the Rewriter API is only available to top-level windows and to their
same-origin iframes. Access to the API can be delegated to cross-origin iframes
using the Permission Policy `allow=""` attribute:
```html
```
The Rewriter API isn't available in Web Workers. This is due to the complexity of
establishing a responsible document for each worker, in order to check the
Permissions Policy status.
## Engage and share feedback
The Writer and Rewriter APIs are under active discussion and subject to change
in the future. If you try this API and have feedback, we'd love to hear it.
- [Read the explainer](https://github.com/explainers-by-googlers/writing-assistance-apis/), raise questions and participate in discussion.
- [Review Chrome Status](https://chromestatus.com/feature/5112320150470656) to see the implementation of the Rewriter API in Chrome.
- If you have feedback on Chrome's implementation, file a [Chromium bug](https://new.crbug.com/).
- [Join the early preview program](https://developer.chrome.com/docs/ai/join-epp.md.txt) for an early look at new APIs and access to our mailing list.
Discover all of the [built-in AI APIs](https://developer.chrome.com/docs/ai/built-in-apis.md.txt) which use
foundation and expert models in the browser.