lee/puter
1
0
Fork 0
mirror of https://github.com/HeyPuter/puter.git synced 2026-05-04 16:40:41 +00:00
Code Issues Actions 8 Packages Projects Releases Wiki Activity

Fix AI types (#2100)

Browse Source 
* Remove constructor and global methods

* Fix streaming type and stricter type

* Add missing chat options param

* Fix messages type

* Fix img2txt types

* Fix txt2img types

* remove unused

* Fix test mode in txt2img

* Fix txt2vid types

* Fix speech2txt types

* Fix speech2speech types

* Fix txt2speech types

* Add missing chat types

* Fix missing txt2speech types
This commit is contained in:
Reynaldi Chernando
2025-12-09 05:06:36 +07:00
committed by GitHub
parent f09cac99eb
commit 7b5d90d678
1 changed files with 107 additions and 34 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/puter-js/types/modules/ai.d.ts
Vendored
+107 -34
View File
@@ -3,34 +3,53 @@ export type AIMessageContent = string | { image_url?: { url: string } } | Record
export interface ChatMessage {
role?: string;
content: AIMessageContent | AIMessageContent[];
[key: string]: unknown;
tool_calls?: ToolCall[];
}
export interface ToolCall {
id: string;
function: { name: string, arguments: string };
}
export interface ChatOptions {
model?: string;
temperature?: number;
max_tokens?: number;
stream?: boolean;
vision?: boolean;
[key: string]: unknown;
driver?: string;
tools?: unknown;
response?: unknown;
reasoning?: unknown;
reasoning_effort?: string;
text?: unknown;
verbosity?: unknown;
}
export interface StreamingChatOptions extends ChatOptions {
stream: boolean;
}
export interface ChatResponse {
message?: ChatMessage;
choices?: unknown;
[key: string]: unknown;
}
export interface ChatResponseChunk {
text?: string;
[key: string]: unknown;
reasoning?: string;
}
export interface Img2TxtOptions {
source?: string | File | Blob;
provider?: string;
testMode?: boolean;
[key: string]: unknown;
model?: string;
pages?: number[];
includeImageBase64?: boolean;
imageLimit?: number;
imageMinSize?: number;
bboxAnnotationFormat?: string;
documentAnnotationFormat?: string;
}
export interface Txt2ImgOptions {
@@ -39,18 +58,38 @@ export interface Txt2ImgOptions {
quality?: string;
input_image?: string;
input_image_mime_type?: string;
[key: string]: unknown;
driver?: string;
provider?: string;
service?: string;
ratio?: { w: number; h: number };
width?: number;
height?: number;
aspect_ratio?: string;
steps?: number;
seed?: number;
negative_prompt?: string;
n?: number;
image_url?: string;
image_base64?: string;
mask_image_url?: string;
mask_image_base64?: string;
prompt_strength?: number;
disable_safety_checker?: boolean;
response_format?: string;
}
export interface Txt2VidOptions {
prompt?: string;
model?: string;
duration?: number;
seconds?: number;
width?: number;
height?: number;
fps?: number;
steps?: number;
[key: string]: unknown;
driver?: string;
provider?: string;
service?: string;
}
export interface Txt2SpeechOptions {
@@ -61,14 +100,9 @@ export interface Txt2SpeechOptions {
provider?: string;
model?: string;
response_format?: string;
[key: string]: unknown;
}
export interface Txt2SpeechCallable {
(text: string, options?: Txt2SpeechOptions): Promise<HTMLAudioElement>;
(text: string, language?: string, voice?: string, engine?: string): Promise<HTMLAudioElement>;
listEngines: (options?: string | Record<string, unknown>) => Promise<unknown>;
listVoices: (options?: string | Record<string, unknown>) => Promise<unknown>;
output_format?: string;
instructions?: string;
voice_settings?: Record<string, unknown>;
}
export interface Speech2TxtOptions {
@@ -80,7 +114,13 @@ export interface Speech2TxtOptions {
prompt?: string;
stream?: boolean;
translate?: boolean;
[key: string]: unknown;
temperature?: number;
logprobs?: boolean;
timestamp_granularities?: string[];
chunking_strategy?: string;
known_speaker_names?: string[];
known_speaker_references?: string[];
extra_body?: Record<string, unknown>;
}
export interface Speech2SpeechOptions {
@@ -88,36 +128,69 @@ export interface Speech2SpeechOptions {
file?: string | File | Blob;
provider?: string;
model?: string;
modelId?: string;
model_id?: string;
voice?: string;
[key: string]: unknown;
voiceId?: string;
voice_id?: string;
output_format?: string;
outputFormat?: string;
voice_settings?: Record<string, unknown>;
voiceSettings?: Record<string, unknown>;
file_format?: string;
fileFormat?: string;
remove_background_noise?: boolean;
removeBackgroundNoise?: boolean;
optimize_streaming_latency?: number;
optimizeStreamingLatency?: number;
enable_logging?: boolean;
enableLogging?: boolean;
}
export class AI {
constructor (context: { authToken?: string; APIOrigin: string; appID?: string });
setAuthToken (authToken: string): void;
setAPIOrigin (APIOrigin: string): void;
listModels (provider?: string): Promise<Record<string, unknown>[]>;
listModelProviders (): Promise<string[]>;
chat (prompt: string, options?: ChatOptions): Promise<ChatResponse>;
chat (prompt: string, imageURL: string | File, options?: ChatOptions): Promise<ChatResponse>;
chat (messages: ChatMessage[], options?: ChatOptions): Promise<ChatResponse>;
chat (prompt: string, options: ChatOptions & { stream: true }): AsyncIterable<ChatResponseChunk>;
chat (prompt: string, imageURL: string | File, options: ChatOptions & { stream: true }): AsyncIterable<ChatResponseChunk>;
chat (messages: ChatMessage[], options: ChatOptions & { stream: true }): AsyncIterable<ChatResponseChunk>;
chat (...args: unknown[]): Promise<ChatResponse> | AsyncIterable<ChatResponseChunk>;
chat (prompt: string, testMode?: boolean): Promise<ChatResponse>;
chat (prompt: string, options: ChatOptions, testMode?: boolean): Promise<ChatResponse>;
chat (prompt: string, imageURL: string | File, testMode?: boolean): Promise<ChatResponse>;
chat (prompt: string, imageURLArray: string[], testMode?: boolean): Promise<ChatResponse>;
chat (prompt: string, imageURL: string | File, options: ChatOptions, testMode?: boolean): Promise<ChatResponse>;
chat (prompt: string, imageURLArray: string[], options: ChatOptions, testMode?: boolean): Promise<ChatResponse>;
chat (prompt: string, options: StreamingChatOptions, testMode?: boolean): AsyncIterable<ChatResponseChunk>;
chat (prompt: string, imageURL: string | File, options: StreamingChatOptions, testMode?: boolean): AsyncIterable<ChatResponseChunk>;
chat (prompt: string, imageURLArray: string[], options: StreamingChatOptions, testMode?: boolean): AsyncIterable<ChatResponseChunk>;
chat (messages: ChatMessage[], testMode?: boolean): Promise<ChatResponse>;
chat (messages: ChatMessage[], options: ChatOptions, testMode?: boolean): Promise<ChatResponse>;
chat (messages: ChatMessage[], options: StreamingChatOptions, testMode?: boolean): AsyncIterable<ChatResponseChunk>;
img2txt (source: string | File | Blob, testMode?: boolean): Promise<string>;
img2txt (source: string | File | Blob, options: Img2TxtOptions, testMode?: boolean): Promise<string>;
img2txt (options: Img2TxtOptions, testMode?: boolean): Promise<string>;
img2txt (source: string | File | Blob | Img2TxtOptions, testMode?: boolean): Promise<string>;
txt2img (prompt: string, testMode?: boolean): Promise<HTMLImageElement>;
txt2img (prompt: string, options: Txt2ImgOptions): Promise<HTMLImageElement>;
txt2img (options: Txt2ImgOptions, testMode?: boolean): Promise<HTMLImageElement>;
txt2vid (prompt: string, testMode?: boolean): Promise<HTMLVideoElement>;
txt2vid (prompt: string, options: Txt2VidOptions): Promise<HTMLVideoElement>;
speech2txt (source: string | File | Blob | Speech2TxtOptions, options?: Speech2TxtOptions): Promise<string | Record<string, unknown>>;
speech2speech (source: string | File | Blob | Speech2SpeechOptions, options?: Speech2SpeechOptions): Promise<Record<string, unknown>>;
txt2vid (options: Txt2VidOptions, testMode?: boolean): Promise<HTMLVideoElement>;
txt2speech: Txt2SpeechCallable;
speech2txt (source: string | File | Blob, testMode?: boolean): Promise<string | Record<string, unknown>>;
speech2txt (source: string | File | Blob, options: Speech2TxtOptions, testMode?: boolean): Promise<string | Record<string, unknown>>;
speech2txt (options: Speech2TxtOptions, testMode?: boolean): Promise<string | Record<string, unknown>>;
speech2speech (source: string | File | Blob, testMode?: boolean): Promise<HTMLAudioElement>;
speech2speech (source: string | File | Blob, options: Speech2SpeechOptions, testMode?: boolean): Promise<HTMLAudioElement>;
speech2speech (options: Speech2SpeechOptions, testMode?: boolean): Promise<HTMLAudioElement>;
txt2speech (text: string, testMode?: boolean): Promise<HTMLAudioElement>;
txt2speech (text: string, options: Txt2SpeechOptions, testMode?: boolean): Promise<HTMLAudioElement>;
txt2speech (text: string, language: string, testMode?: boolean): Promise<HTMLAudioElement>;
txt2speech (text: string, language: string, voice: string, testMode?: boolean): Promise<HTMLAudioElement>;
txt2speech (text: string, language: string, voice: string, engine: string, testMode?: boolean): Promise<HTMLAudioElement>;
}
// NOTE: AI responses contain provider-specific payloads that are not fully typed here because