WPT details: The API surface is reasonably well tested with mocked models, but production model downloading and non-deterministic outputs are not fully tested at the web-platform-tests layer. The explainer discusses this in https://github.com/WICG/writing-assistance-apis/blob/main/README.md#specifications-and-tests.
The Writer API can be used for writing new material given a writing task prompt, backed by an on-device AI language model. Developers will be able to use this API to generate textual explanations of structured data, composing a post about a product based on reviews or product description, expanding on pro and con lists into full views and more. An enterprise policy (GenAILocalFoundationalModelSettings) is available to disable the underlying model downloading which would render this API unavailable.
Browsers and operating systems are increasingly expected to gain access to a language model. By exposing this built-in model, we avoid every website needing to download their own multi-gigabyte language model, or send input text to third-party APIs. The writer API in particular exposes a high-level API for interfacing with a language model in order to produce new output for a variety of use cases [1], in a way that does not depend on the specific language model in question. [1]: https://github.com/explainers-by-googlers/writing-assistance-apis/blob/main/README.md#writer-api
Samples: https://chrome.dev/web-ai-demos/writer-rewriter-api-playground https://ai.etiennenoel.com/writer-api
Explainers: https://github.com/explainers-by-googlers/writing-assistance-apis/blob/main/README.md