Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Remove mention of returning data URLs from sync API #330

Merged
merged 3 commits into from
Oct 30, 2024
Merged
Changes from 1 commit
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
21 changes: 2 additions & 19 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -68,10 +68,6 @@ console.log(output.blob()); // Blob
> A model that outputs file data returns a `FileOutput` object by default. This is an implementation
> of `ReadableStream` that returns the file contents. It has a `.blob()` method for accessing a
> `Blob` representation and a `.url()` method that will return the underlying data-source.
>
> **This data source can be either a remote URL or a data-uri with base64 encoded data. Check
> out the documentation on [creating a prediction](https://replicate.com/docs/topics/predictions/create-a-prediction)
> for more information.**
aron marked this conversation as resolved.
Show resolved Hide resolved

You can also run a model in the background:

Expand Down Expand Up @@ -388,19 +384,6 @@ which is optimized for low latency requests to file models like `black-forest-la
`black-forest-labs/flux-schnell`. When creating a prediction this will hold a connection open to the
server and return a `FileObject` containing the generated file as quickly as possible.

> [!NOTE]
> In this mode the `url()` method on the `FileObject` may refer to either a remote URL or
> base64 encoded data-uri. The latter is an optimization we make on certain models to deliver
> the files faster to the client.
>
> If you need the prediction URLs for whatever reason you can opt out of the sync mode by
> passing `wait: { "type": "poll" }` to the `run()` method.
>
> ```js
> const output = await replicate.run(model, { input, wait: { type: "poll" } });
> output.url() // URL<https://...>
> ```

### `replicate.stream`

Run a model and stream its output. Unlike [`replicate.prediction.create`](#replicatepredictionscreate), this method returns only the prediction output rather than the entire prediction object.
Expand Down Expand Up @@ -1238,7 +1221,7 @@ const [output] = await replicate.run("black-forest-labs/flux-schnell", {
input: { prompt: "astronaut riding a rocket like a horse" }
});

// To access the file URL (or data-uri):
// To access the file URL:
console.log(output.url()); //=> "http://example.com"

// To write the file to disk:
Expand All @@ -1261,7 +1244,7 @@ const replicate = new Replicate({ useFileOutput: false });

| method | returns | description |
| -------------------- | ------ | ------------------------------------------------------------ |
| `url()` | string | A `URL` object representing the HTTP URL or data-uri |
aron marked this conversation as resolved.
Show resolved Hide resolved
| `url()` | string | A `URL` object pointing to the underlying data source |
| `blob()` | object | A `Blob` instance containing the binary file |

## Troubleshooting
Expand Down
Loading