Merge branch 'main' into add_sam3d

Author: merveenoyan

README.md

@@ -98,7 +98,7 @@ You can run our packages with vanilla JS, without any bundler, by using a CDN or
 
 ```html
 <script type="module">
-    import { InferenceClient } from 'https://cdn.jsdelivr.net/npm/@huggingface/[email protected].4/+esm';
+    import { InferenceClient } from 'https://cdn.jsdelivr.net/npm/@huggingface/[email protected].5/+esm';
     import { createRepo, commit, deleteRepo, listFiles } from "https://cdn.jsdelivr.net/npm/@huggingface/[email protected]/+esm";
 </script>
 ```

packages/hub/src/lib/collection-info.spec.ts

@@ -31,17 +31,17 @@ describe("collectionInfo", () => {
 					_id: "690df2a467ea25a1a346d0ae",
 					author: "huggingfacejs",
 					datasetsServerInfo: {
-						formats: ["imagefolder"],
-						libraries: ["datasets", "mlcroissant"],
-						modalities: ["audio", "image"],
-						numRows: 38,
-						viewer: "viewer",
+						formats: expect.any(Array),
+						libraries: expect.any(Array),
+						modalities: expect.any(Array),
+						numRows: expect.any(Number),
+						viewer: expect.any(String),
 					},
 					downloads: expect.any(Number),
 					gated: false,
 					id: "huggingfacejs/tasks",
 					isLikedByUser: false,
-					lastModified: "2025-05-21T10:59:37.000Z",
+					lastModified: expect.any(String),
 					likes: expect.any(Number),
 					position: 0,
 					private: false,
@@ -57,8 +57,7 @@ describe("collectionInfo", () => {
 					numberItems: 5,
 					owner: {
 						_id: "6414d83b385a75d7790d5a58",
-						avatarUrl:
-							"https://cdn-avatars.huggingface.co/v1/production/uploads/1679315631188-61d2f90c3c2083e1c08af22d.png",
+						avatarUrl: expect.any(String),
 						followerCount: expect.any(Number),
 						fullname: "Huggingface.js",
 						isEnterprise: false,
@@ -83,8 +82,7 @@ describe("collectionInfo", () => {
 					author: "huggingfacejs",
 					authorData: {
 						_id: "6414d83b385a75d7790d5a58",
-						avatarUrl:
-							"https://cdn-avatars.huggingface.co/v1/production/uploads/1679315631188-61d2f90c3c2083e1c08af22d.png",
+						avatarUrl: expect.any(String),
 						followerCount: expect.any(Number),
 						fullname: "Huggingface.js",
 						isEnterprise: false,
@@ -98,12 +96,12 @@ describe("collectionInfo", () => {
 					colorTo: "green",
 					createdAt: "2023-03-17T21:33:16.000Z",
 					emoji: "🌎",
+					featured: false,
 					id: "huggingfacejs/push-model-from-web",
 					isLikedByUser: false,
-					lastModified: "2023-04-01T15:29:38.000Z",
-					likes: expect.any(Number),
+					lastModified: expect.any(String),
+					likes: 1,
 					pinned: false,
-					featured: false,
 					position: 2,
 					private: false,
 					repoType: "space",
@@ -141,8 +139,7 @@ describe("collectionInfo", () => {
 					author: "huggingfacejs",
 					authorData: {
 						_id: "6414d83b385a75d7790d5a58",
-						avatarUrl:
-							"https://cdn-avatars.huggingface.co/v1/production/uploads/1679315631188-61d2f90c3c2083e1c08af22d.png",
+						avatarUrl: expect.any(String),
 						followerCount: expect.any(Number),
 						fullname: "Huggingface.js",
 						isEnterprise: false,
@@ -157,7 +154,7 @@ describe("collectionInfo", () => {
 					gated: false,
 					id: "huggingfacejs/test-model",
 					isLikedByUser: false,
-					lastModified: "2025-11-07T13:25:23.000Z",
+					lastModified: expect.any(String),
 					likes: expect.any(Number),
 					position: 4,
 					private: false,

packages/hub/src/lib/list-collections.spec.ts

@@ -41,17 +41,17 @@ describe("listCollections", () => {
 						_id: "690df2a467ea25a1a346d0ae",
 						author: "huggingfacejs",
 						datasetsServerInfo: {
-							formats: ["imagefolder"],
-							libraries: ["datasets", "mlcroissant"],
-							modalities: ["audio", "image"],
-							numRows: 38,
-							viewer: "viewer",
+							formats: expect.any(Array),
+							libraries: expect.any(Array),
+							modalities: expect.any(Array),
+							numRows: expect.any(Number),
+							viewer: expect.any(String),
 						},
 						downloads: expect.any(Number),
 						gated: false,
 						id: "huggingfacejs/tasks",
 						isLikedByUser: false,
-						lastModified: "2025-05-21T10:59:37.000Z",
+						lastModified: expect.any(String),
 						likes: expect.any(Number),
 						position: 0,
 						private: false,
@@ -63,12 +63,11 @@ describe("listCollections", () => {
 						description: "Only used in E2E tests",
 						id: "690df2897fa1945492b8cf42",
 						isUpvotedByUser: false,
-						lastUpdated: "2025-11-07T13:24:26.277Z",
+						lastUpdated: expect.any(String),
 						numberItems: 5,
 						owner: {
 							_id: "6414d83b385a75d7790d5a58",
-							avatarUrl:
-								"https://cdn-avatars.huggingface.co/v1/production/uploads/1679315631188-61d2f90c3c2083e1c08af22d.png",
+							avatarUrl: expect.any(String),
 							followerCount: expect.any(Number),
 							fullname: "Huggingface.js",
 							isEnterprise: false,
@@ -93,8 +92,7 @@ describe("listCollections", () => {
 						author: "huggingfacejs",
 						authorData: {
 							_id: "6414d83b385a75d7790d5a58",
-							avatarUrl:
-								"https://cdn-avatars.huggingface.co/v1/production/uploads/1679315631188-61d2f90c3c2083e1c08af22d.png",
+							avatarUrl: expect.any(String),
 							followerCount: expect.any(Number),
 							fullname: "Huggingface.js",
 							isEnterprise: false,
@@ -108,14 +106,14 @@ describe("listCollections", () => {
 						colorTo: "green",
 						createdAt: "2023-03-17T21:33:16.000Z",
 						emoji: "🌎",
+						featured: false,
 						id: "huggingfacejs/push-model-from-web",
 						isLikedByUser: false,
-						lastModified: "2023-04-01T15:29:38.000Z",
-						likes: expect.any(Number),
+						lastModified: expect.any(String),
+						likes: 1,
 						pinned: false,
 						position: 2,
 						private: false,
-						featured: false,
 						repoType: "space",
 						runtime: {
 							hardware: {

packages/inference/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@huggingface/inference",
-	"version": "4.13.4",
+	"version": "4.13.5",
 	"packageManager": "[email protected]",
 	"license": "MIT",
 	"author": "Hugging Face and Tim Mikeladze <[email protected]>",

packages/inference/src/lib/getProviderHelper.ts

@@ -28,6 +28,8 @@ import type {
 	ImageToImageTaskHelper,
 	ImageToTextTaskHelper,
 	ImageToVideoTaskHelper,
+	ImageTextToImageTaskHelper,
+	ImageTextToVideoTaskHelper,
 	ObjectDetectionTaskHelper,
 	QuestionAnsweringTaskHelper,
 	SentenceSimilarityTaskHelper,
@@ -276,6 +278,14 @@ export function getProviderHelper(
 	provider: InferenceProviderOrPolicy,
 	task: "image-to-video"
 ): ImageToVideoTaskHelper & TaskProviderHelper;
+export function getProviderHelper(
+	provider: InferenceProviderOrPolicy,
+	task: "image-text-to-image"
+): ImageTextToImageTaskHelper & TaskProviderHelper;
+export function getProviderHelper(
+	provider: InferenceProviderOrPolicy,
+	task: "image-text-to-video"
+): ImageTextToVideoTaskHelper & TaskProviderHelper;
 export function getProviderHelper(
 	provider: InferenceProviderOrPolicy,
 	task: "sentence-similarity"

packages/inference/src/package.ts

@@ -1,3 +1,3 @@
 // Generated file from package.json. Issues importing JSON directly when publishing on commonjs/ESM - see https://github.com/microsoft/TypeScript/issues/51783
-export const PACKAGE_VERSION = "4.13.4";
+export const PACKAGE_VERSION = "4.13.5";
 export const PACKAGE_NAME = "@huggingface/inference";

packages/inference/src/providers/providerHelper.ts

@@ -19,6 +19,8 @@ import type {
 	ImageToTextInput,
 	ImageToTextOutput,
 	ImageToVideoInput,
+	ImageTextToImageInput,
+	ImageTextToVideoInput,
 	ObjectDetectionInput,
 	ObjectDetectionOutput,
 	QuestionAnsweringInput,
@@ -54,6 +56,8 @@ import { toArray } from "../utils/toArray.js";
 import type { ImageToImageArgs } from "../tasks/cv/imageToImage.js";
 import type { AutomaticSpeechRecognitionArgs } from "../tasks/audio/automaticSpeechRecognition.js";
 import type { ImageToVideoArgs } from "../tasks/cv/imageToVideo.js";
+import type { ImageTextToImageArgs } from "../tasks/cv/imageTextToImage.js";
+import type { ImageTextToVideoArgs } from "../tasks/cv/imageTextToVideo.js";
 import type { ImageSegmentationArgs } from "../tasks/cv/imageSegmentation.js";
 
 /**
@@ -159,6 +163,18 @@ export interface ImageToVideoTaskHelper {
 	preparePayloadAsync(args: ImageToVideoArgs): Promise<RequestArgs>;
 }
 
+export interface ImageTextToImageTaskHelper {
+	getResponse(response: unknown, url?: string, headers?: HeadersInit): Promise<Blob>;
+	preparePayload(params: BodyParams<ImageTextToImageInput & BaseArgs>): Record<string, unknown>;
+	preparePayloadAsync(args: ImageTextToImageArgs): Promise<RequestArgs>;
+}
+
+export interface ImageTextToVideoTaskHelper {
+	getResponse(response: unknown, url?: string, headers?: HeadersInit): Promise<Blob>;
+	preparePayload(params: BodyParams<ImageTextToVideoInput & BaseArgs>): Record<string, unknown>;
+	preparePayloadAsync(args: ImageTextToVideoArgs): Promise<RequestArgs>;
+}
+
 export interface ImageSegmentationTaskHelper {
 	getResponse(response: unknown, url?: string, headers?: HeadersInit): Promise<ImageSegmentationOutput>;
 	preparePayload(params: BodyParams<ImageSegmentationInput & BaseArgs>): Record<string, unknown> | BodyInit;

packages/inference/src/snippets/getInferenceSnippets.ts

@@ -96,7 +96,10 @@ const HF_PYTHON_METHODS: Partial<Record<WidgetType, string>> = {
 	"image-classification": "image_classification",
 	"image-segmentation": "image_segmentation",
 	"image-to-image": "image_to_image",
+	"image-to-video": "image_to_video",
 	"image-to-text": "image_to_text",
+	"image-text-to-image": "image_text_to_image",
+	"image-text-to-video": "image_text_to_video",
 	"object-detection": "object_detection",
 	"question-answering": "question_answering",
 	"sentence-similarity": "sentence_similarity",
@@ -390,7 +393,9 @@ const snippets: Partial<
 	"fill-mask": snippetGenerator("basic"),
 	"image-classification": snippetGenerator("basicImage"),
 	"image-segmentation": snippetGenerator("basicImage"),
+	"image-text-to-image": snippetGenerator("imageToImage", prepareImageToImageInput),
 	"image-text-to-text": snippetGenerator("conversational"),
+	"image-text-to-video": snippetGenerator("imageToVideo", prepareImageToImageInput),
 	"image-to-image": snippetGenerator("imageToImage", prepareImageToImageInput),
 	"image-to-text": snippetGenerator("basicImage"),
 	"image-to-video": snippetGenerator("imageToVideo", prepareImageToImageInput),

packages/inference/src/tasks/cv/imageTextToImage.ts

@@ -0,0 +1,22 @@
+import type { ImageTextToImageInput } from "@huggingface/tasks";
+import { resolveProvider } from "../../lib/getInferenceProviderMapping.js";
+import { getProviderHelper } from "../../lib/getProviderHelper.js";
+import type { BaseArgs, Options } from "../../types.js";
+import { innerRequest } from "../../utils/request.js";
+
+export type ImageTextToImageArgs = BaseArgs & ImageTextToImageInput;
+
+/**
+ * This task takes an image and text input and outputs a new generated image.
+ * Recommended model: black-forest-labs/FLUX.2-dev
+ */
+export async function imageTextToImage(args: ImageTextToImageArgs, options?: Options): Promise<Blob> {
+	const provider = await resolveProvider(args.provider, args.model, args.endpointUrl);
+	const providerHelper = getProviderHelper(provider, "image-text-to-image");
+	const payload = await providerHelper.preparePayloadAsync(args);
+	const { data: res, requestContext } = await innerRequest<Blob>(payload, providerHelper, {
+		...options,
+		task: "image-text-to-image",
+	});
+	return providerHelper.getResponse(res, requestContext.url, requestContext.info.headers as Record<string, string>);
+}

packages/inference/src/tasks/cv/imageTextToVideo.ts

@@ -0,0 +1,22 @@
+import type { ImageTextToVideoInput } from "@huggingface/tasks";
+import { resolveProvider } from "../../lib/getInferenceProviderMapping.js";
+import { getProviderHelper } from "../../lib/getProviderHelper.js";
+import type { BaseArgs, Options } from "../../types.js";
+import { innerRequest } from "../../utils/request.js";
+
+export type ImageTextToVideoArgs = BaseArgs & ImageTextToVideoInput;
+
+/**
+ * This task takes an image and text input and outputs a generated video.
+ * Recommended model: Lightricks/LTX-Video
+ */
+export async function imageTextToVideo(args: ImageTextToVideoArgs, options?: Options): Promise<Blob> {
+	const provider = await resolveProvider(args.provider, args.model, args.endpointUrl);
+	const providerHelper = getProviderHelper(provider, "image-text-to-video");
+	const payload = await providerHelper.preparePayloadAsync(args);
+	const { data: res, requestContext } = await innerRequest<Blob>(payload, providerHelper, {
+		...options,
+		task: "image-text-to-video",
+	});
+	return providerHelper.getResponse(res, requestContext.url, requestContext.info.headers as Record<string, string>);
+}