Prompt API

1. Introduction

TODO

2. Dependencies

This specification depends on the Infra Standard. [INFRA]

As with the rest of the web platform, human languages are identified in these APIs by BCP 47 language tags, such as " ja ", " en-US ", " sr-Cyrl ", or " de-CH-1901-x-phonebk-extended ". The specific algorithms used for validation, canonicalization, and language tag matching are those from the ECMAScript Internationalization API Specification , which in turn defers some of its processing to Unicode Locale Data Markup Language (LDML) . [BCP47] [ECMA-402] [UTS35] .

These APIs are part of a family of APIs expected to be powered by machine learning models, which share common API surface idioms and specification patterns. Currently, the specification text for these shared parts lives in Writing Assistance APIs § 5 Shared infrastructure , and the common privacy and security considerations are discussed in Writing Assistance APIs § 6 Privacy considerations and Writing Assistance APIs § 7 Security considerations . Implementing these APIs requires implementing that shared infrastructure, and conforming to those privacy and security considerations. But it does not require implementing or exposing the actual writing assistance APIs. [WRITING-ASSISTANCE-APIS]

3. The API

[Exposed=Window, SecureContext]
interface LanguageModel : EventTarget {
  static Promise<LanguageModel> create(optional LanguageModelCreateOptions options = {});
  static Promise<Availability> availability(optional LanguageModelCreateCoreOptions options = {});
  static Promise<LanguageModelParams?> params();
  // These will throw "NotSupportedError" DOMExceptions if role = "system"
  Promise<DOMString> prompt(
    LanguageModelPrompt input,
    optional LanguageModelPromptOptions options = {}
  );
  ReadableStream promptStreaming(
    LanguageModelPrompt input,
    optional LanguageModelPromptOptions options = {}
  );
  Promise<undefined> append(
    LanguageModelPrompt input,
    optional LanguageModelAppendOptions options = {}
  );
  Promise<double> measureInputUsage(
    LanguageModelPrompt input,
    optional LanguageModelPromptOptions options = {}
  );
  readonly attribute double inputUsage;
  readonly attribute unrestricted double inputQuota;
  attribute EventHandler onquotaoverflow;
  readonly attribute unsigned long topK;
  readonly attribute float temperature;
  Promise<LanguageModel> clone(optional LanguageModelCloneOptions options = {});
  undefined destroy();
};
[Exposed=Window, SecureContext]
interface LanguageModelParams {
  readonly attribute unsigned long defaultTopK;
  readonly attribute unsigned long maxTopK;
  readonly attribute float defaultTemperature;
  readonly attribute float maxTemperature;
};
dictionary LanguageModelCreateCoreOptions {
  // Note: these two have custom out-of-range handling behavior, not in the IDL layer.
  // They are unrestricted double so as to allow +Infinity without failing.
  unrestricted double topK;
  unrestricted double temperature;
  sequence<LanguageModelExpected> expectedInputs;
  sequence<LanguageModelExpected> expectedOutputs;
};
dictionary LanguageModelCreateOptions : LanguageModelCreateCoreOptions {
  AbortSignal signal;
  CreateMonitorCallback monitor;
  sequence<LanguageModelMessage> initialPrompts;
};
dictionary LanguageModelPromptOptions {
  object responseConstraint;
  AbortSignal signal;
};
dictionary LanguageModelAppendOptions {
  AbortSignal signal;
};
dictionary LanguageModelCloneOptions {
  AbortSignal signal;
};
dictionary LanguageModelExpected {
  required LanguageModelMessageType type;
  sequence<DOMString> languages;
};
// The argument to the prompt() method and others like it
typedef (
  sequence<LanguageModelMessage>
  // Shorthand for `[{ role: "user", content: [{ type: "text", value: providedValue }] }]`
  or DOMString
) LanguageModelPrompt;
dictionary LanguageModelMessage {
  required LanguageModelMessageRole role;
  // The DOMString branch is shorthand for `[{ type: "text", value: providedValue }]`
  required (DOMString or sequence<LanguageModelMessageContent>) content;
};
dictionary LanguageModelMessageContent {
  required LanguageModelMessageType type;
  required LanguageModelMessageValue value;
};
enum LanguageModelMessageRole { "system", "user", "assistant" };
enum LanguageModelMessageType { "text", "image", "audio" };
typedef (
  ImageBitmapSource
  or AudioBuffer
  or BufferSource
  or DOMString
) LanguageModelMessageValue;

3.1. Prompt processing

This will be incorporated into a proper part of the specification later. For now, we’re just writing out this algorithm as a full spec, since it’s complicated.

3.2. Permissions policy integration

Access to the prompt API is gated behind the policy-controlled feature " language-model ", which has a default allowlist of 'self' .

4. Privacy considerations

Please see Writing Assistance APIs § 6 Privacy considerations for a discussion of privacy considerations for the prompt API. That text was written to apply to all APIs sharing the same infrastructure, as noted in § 2 Dependencies .

5. Security considerations

Please see Writing Assistance APIs § 7 Security considerations for a discussion of security considerations for the prompt API. That text was written to apply to all APIs sharing the same infrastructure, as noted in § 2 Dependencies .