@huggingface/tasks - npm Package Compare versions

Comparing version 0.2.1 to 0.2.2

dist/index.d.ts

@@ -8,3 +8,3 @@ /**
 declare enum ModelLibrary {
-    "adapter-transformers" = "Adapter Transformers",
+    "adapter-transformers" = "Adapters",
     "allennlp" = "allenNLP",
@@ -452,3 +452,3 @@ "asteroid" = "Asteroid",
  * Mapping from library name (excluding Transformers) to its supported tasks.
- * Inference API should be disabled for all other (library, task) pairs beyond this mapping.
+ * Inference Endpoints (serverless) should be disabled for all other (library, task) pairs beyond this mapping.
  * As an exception, we assume Transformers supports all inference tasks.
@@ -617,3 +617,3 @@ * This mapping is generated automatically by "python-api-export-tasks" action in huggingface/api-inference-community repo upon merge.
     /**
-     * Parameters that will be used by the widget when calling Inference API
+     * Parameters that will be used by the widget when calling Inference Endpoints (serverless)
      * https://huggingface.co/docs/api-inference/detailed_parameters
@@ -737,2 +737,3 @@ *
     id: PipelineType;
+    canonicalId?: PipelineType;
     isPlaceholder?: boolean;
@@ -739,0 +740,0 @@ label: string;

package.json

 {
   "name": "@huggingface/tasks",
   "packageManager": "pnpm@8.10.5",
-  "version": "0.2.1",
+  "version": "0.2.2",
   "description": "List of ML tasks for huggingface.co/tasks",
@@ -10,4 +10,4 @@ "repository": "https://github.com/huggingface/huggingface.js.git",
   },
-  "main": "./dist/index.js",
-  "module": "./dist/index.mjs",
+  "main": "./dist/index.cjs",
+  "module": "./dist/index.js",
   "types": "./dist/index.d.ts",
@@ -17,7 +17,8 @@ "exports": {
       "types": "./dist/index.d.ts",
-      "require": "./dist/index.js",
-      "import": "./dist/index.mjs"
+      "require": "./dist/index.cjs",
+      "import": "./dist/index.js"
     }
   },
   "source": "src/index.ts",
+  "type": "module",
   "files": [
@@ -35,3 +36,6 @@ "dist",
   "license": "MIT",
-  "devDependencies": {},
+  "devDependencies": {
+    "@types/node": "^20.11.5",
+    "quicktype-core": "https://github.com/huggingface/quicktype/raw/pack-18.0.15/packages/quicktype-core/quicktype-core-18.0.15.tgz"
+  },
   "scripts": {
@@ -42,5 +46,6 @@ "lint": "eslint --quiet --fix --ext .cjs,.ts .",
     "format:check": "prettier --check .",
-    "build": "tsup src/index.ts --format cjs,esm --clean --dts",
-    "check": "tsc"
+    "build": "tsup src/index.ts --format cjs,esm --clean --dts && pnpm run inference-codegen",
+    "check": "tsc",
+    "inference-codegen": "tsx scripts/inference-codegen.ts && prettier --write src/tasks/*/inference.ts"
   }
 }

src/library-to-tasks.ts

@@ -6,3 +6,3 @@ import type { ModelLibraryKey } from "./model-libraries";
  * Mapping from library name (excluding Transformers) to its supported tasks.
- * Inference API should be disabled for all other (library, task) pairs beyond this mapping.
+ * Inference Endpoints (serverless) should be disabled for all other (library, task) pairs beyond this mapping.
  * As an exception, we assume Transformers supports all inference tasks.
@@ -9,0 +9,0 @@ * This mapping is generated automatically by "python-api-export-tasks" action in huggingface/api-inference-community repo upon merge.

src/library-ui-elements.ts

@@ -38,7 +38,7 @@ import type { ModelData } from "./model-data";
 
-const adapter_transformers = (model: ModelData) => [
-	`from transformers import ${model.config?.adapter_transformers?.model_class}
+const adapters = (model: ModelData) => [
+	`from adapters import AutoAdapterModel
 
-model = ${model.config?.adapter_transformers?.model_class}.from_pretrained("${model.config?.adapter_transformers?.model_name}")
-model.load_adapter("${model.id}", source="hf")`,
+model = AutoAdapterModel.from_pretrained("${model.config?.adapter_transformers?.model_name}")
+model.load_adapter("${model.id}", set_active=True)`,
 ];
@@ -545,3 +545,3 @@
 
-const sentis = (model: ModelData) => [
+const sentis = (/* model: ModelData */) => [
 	`string modelName = "[Your model name here].sentis";
@@ -551,3 +551,3 @@ Model model = ModelLoader.Load(Application.streamingAssetsPath + "/" + modelName);
 // Please see provided C# file for more details
-`
+`,
 ];
@@ -582,7 +582,7 @@
 	"adapter-transformers": {
-		btnLabel: "Adapter Transformers",
-		repoName: "adapter-transformers",
-		repoUrl: "https://github.com/Adapter-Hub/adapter-transformers",
-		docsUrl: "https://huggingface.co/docs/hub/adapter-transformers",
-		snippets: adapter_transformers,
+		btnLabel: "Adapters",
+		repoName: "adapters",
+		repoUrl: "https://github.com/Adapter-Hub/adapters",
+		docsUrl: "https://huggingface.co/docs/hub/adapters",
+		snippets: adapters,
 	},
@@ -589,0 +589,0 @@ allennlp: {

src/model-data.ts

@@ -81,3 +81,3 @@ import type { PipelineType } from "./pipelines";
 	/**
-	 * Parameters that will be used by the widget when calling Inference API
+	 * Parameters that will be used by the widget when calling Inference Endpoints (serverless)
 	 * https://huggingface.co/docs/api-inference/detailed_parameters
@@ -84,0 +84,0 @@ *

src/model-libraries.ts

@@ -8,3 +8,3 @@ /**
 export enum ModelLibrary {
-	"adapter-transformers" = "Adapter Transformers",
+	"adapter-transformers" = "Adapters",
 	"allennlp" = "allenNLP",
@@ -11,0 +11,0 @@ "asteroid" = "Asteroid",

src/pipelines.ts

@@ -65,3 +65,3 @@ export const MODALITIES = ["cv", "nlp", "audio", "tabular", "multimodal", "rl", "other"] as const;
 ///  - To determine which widget to show.
-///  - To determine which endpoint of Inference API to use.
+///  - To determine which endpoint of Inference Endpoints to use.
 ///  - As filters at the left of models and datasets page.
@@ -68,0 +68,0 @@ ///

src/tasks/audio-classification/about.md

@@ -29,3 +29,3 @@ ## Use Cases
 
-Emotion recognition is self explanatory. In addition to trying the widgets, you can use the Inference API to perform audio classification. Here is a simple example that uses a [HuBERT](https://huggingface.co/superb/hubert-large-superb-er) model fine-tuned for this task.
+Emotion recognition is self explanatory. In addition to trying the widgets, you can use Inference Endpoints to perform audio classification. Here is a simple example that uses a [HuBERT](https://huggingface.co/superb/hubert-large-superb-er) model fine-tuned for this task.
 
@@ -32,0 +32,0 @@ ```python

src/tasks/audio-to-audio/about.md

@@ -15,3 +15,3 @@ ## Use Cases
 
-Alternatively, you can use the [Inference API](https://huggingface.co/inference-api) to solve this task
+Alternatively, you can use [Inference Endpoints](https://huggingface.co/inference-endpoints) to solve this task
 
@@ -18,0 +18,0 @@ ```python

src/tasks/automatic-speech-recognition/about.md

@@ -21,3 +21,3 @@ ## Use Cases
 
-The Hub contains over [~9,000 ASR models](https://huggingface.co/models?pipeline_tag=automatic-speech-recognition&sort=downloads) that you can use right away by trying out the widgets directly in the browser or calling the models as a service using the Inference API. Here is a simple code snippet to do exactly this:
+The Hub contains over [~9,000 ASR models](https://huggingface.co/models?pipeline_tag=automatic-speech-recognition&sort=downloads) that you can use right away by trying out the widgets directly in the browser or calling the models as a service using Inference Endpoints. Here is a simple code snippet to do exactly this:
 
@@ -87,4 +87,6 @@ ```python
 - [An introduction to SpeechT5, a multi-purpose speech recognition and synthesis model](https://huggingface.co/blog/speecht5)
-- [A guide on Fine-tuning Whisper For Multilingual ASR with 🤗Transformers](https://huggingface.co/blog/fine-tune-whisper)
+- [Fine-tune Whisper For Multilingual ASR with 🤗Transformers](https://huggingface.co/blog/fine-tune-whisper)
 - [Automatic speech recognition task guide](https://huggingface.co/docs/transformers/tasks/asr)
 - [Speech Synthesis, Recognition, and More With SpeechT5](https://huggingface.co/blog/speecht5)
+- [Fine-Tune W2V2-Bert for low-resource ASR with 🤗 Transformers](https://huggingface.co/blog/fine-tune-w2v2-bert)
+- [Speculative Decoding for 2x Faster Whisper Inference](https://huggingface.co/blog/whisper-speculative-decoding)

src/tasks/depth-estimation/data.ts

@@ -31,4 +31,4 @@ import type { TaskDataCustom } from "..";
 		{
-			description: "Strong Depth Estimation model trained on the KITTI dataset.",
-			id: "facebook/dpt-dinov2-large-kitti",
+			description: "Strong Depth Estimation model trained on a big compilation of datasets.",
+			id: "LiheYoung/depth-anything-large-hf",
 		},
@@ -46,5 +46,9 @@ {
 		{
-			description: "An application that can estimate the depth in a given image.",
-			id: "keras-io/Monocular-Depth-Estimation",
+			description: "An application to compare the outputs of different depth estimation models.",
+			id: "LiheYoung/Depth-Anything",
 		},
+		{
+			description: "An application to try state-of-the-art depth estimation.",
+			id: "merve/compare_depth_models",
+		},
 	],
@@ -51,0 +55,0 @@ summary: "Depth estimation is the task of predicting depth of the objects present in an image.",

src/tasks/index.ts

@@ -14,2 +14,3 @@ import { type PipelineType, PIPELINE_DATA } from "../pipelines";
 import imageSegmentation from "./image-segmentation/data";
+import maskGeneration from "./mask-generation/data";
 import objectDetection from "./object-detection/data";
@@ -37,2 +38,3 @@ import depthEstimation from "./depth-estimation/data";
 import zeroShotImageClassification from "./zero-shot-image-classification/data";
+import zeroShotObjectDetection from "./zero-shot-object-detection/data";
 
@@ -136,3 +138,3 @@ import type { ModelLibraryKey } from "../model-libraries";
 	"image-to-video": undefined,
-	"mask-generation": getData("mask-generation", placeholder),
+	"mask-generation": getData("mask-generation", maskGeneration),
 	"multiple-choice": undefined,
@@ -168,3 +170,3 @@ "object-detection": getData("object-detection", objectDetection),
 	"zero-shot-image-classification": getData("zero-shot-image-classification", zeroShotImageClassification),
-	"zero-shot-object-detection": getData("zero-shot-object-detection", placeholder),
+	"zero-shot-object-detection": getData("zero-shot-object-detection", zeroShotObjectDetection),
 	"text-to-3d": getData("text-to-3d", placeholder),
@@ -223,2 +225,3 @@ "image-to-3d": getData("image-to-3d", placeholder),
 	id: PipelineType;
+	canonicalId?: PipelineType;
 	isPlaceholder?: boolean;
@@ -225,0 +228,0 @@ label: string;

src/tasks/mask-generation/data.ts

@@ -6,10 +6,47 @@ import type { TaskDataCustom } from "..";
 	demo: {
-		inputs: [],
-		outputs: [],
+		inputs: [
+			{
+				filename: "mask-generation-input.png",
+				type: "img",
+			},
+		],
+		outputs: [
+			{
+				filename: "mask-generation-output.png",
+				type: "img",
+			},
+		],
 	},
 	metrics: [],
-	models: [],
-	spaces: [],
+	models: [
+		{
+			description: "Small yet powerful mask generation model.",
+			id: "Zigeng/SlimSAM-uniform-50",
+		},
+		{
+			description: "Very strong mask generation model.",
+			id: "facebook/sam-vit-huge",
+		},
+	],
+	spaces: [
+		{
+			description:
+				"An application that combines a mask generation model with an image embedding model for open-vocabulary image segmentation.",
+			id: "SkalskiP/SAM_and_MetaCLIP",
+		},
+		{
+			description: "An application that compares the performance of a large and a small mask generation model.",
+			id: "merve/slimsam",
+		},
+		{
+			description: "An application based on an improved mask generation model.",
+			id: "linfanluntan/Grounded-SAM",
+		},
+		{
+			description: "An application to remove objects from videos using mask generation models.",
+			id: "SkalskiP/SAM_and_ProPainter",
+		},
+	],
 	summary:
-		"Mask generation is creating a binary image that identifies a specific object or region of interest in an input image. Masks are often used in segmentation tasks, where they provide a precise way to isolate the object of interest for further processing or analysis.",
+		"Mask generation is the task of generating masks that identify a specific object or region of interest in a given image. Masks are often used in segmentation tasks, where they provide a precise way to isolate the object of interest for further processing or analysis.",
 	widgetModels: [],
@@ -16,0 +53,0 @@ youtubeId: "",

src/tasks/placeholder/data.ts

@@ -16,4 +16,7 @@ import type { TaskDataCustom } from "..";
 	youtubeId: undefined,
+	/// If this is a subtask, link to the most general task ID
+	/// (eg, text2text-generation is the canonical ID of translation)
+	canonicalId: undefined,
 };
 
 export default taskData;

src/tasks/sentence-similarity/about.md

@@ -11,3 +11,3 @@ ## Use Cases 🔍
 
-You can find and use [hundreds of Sentence Transformers](https://huggingface.co/models?library=sentence-transformers&sort=downloads) models from the Hub by directly using the library, playing with the widgets in the browser or using the Inference API.
+You can find and use [hundreds of Sentence Transformers](https://huggingface.co/models?library=sentence-transformers&sort=downloads) models from the Hub by directly using the library, playing with the widgets in the browser or using Inference Endpoints.
 
@@ -20,3 +20,3 @@ ## Task Variants
 
-You can infer with Passage Ranking models using the [Inference API](https://huggingface.co/inference-api). The Passage Ranking model inputs are a query for which we look for relevancy in the documents and the documents we want to search. The model will return scores according to the relevancy of these documents for the query.
+You can infer with Passage Ranking models using [Inference Endpoints](https://huggingface.co/inference-endpoints). The Passage Ranking model inputs are a query for which we look for relevancy in the documents and the documents we want to search. The model will return scores according to the relevancy of these documents for the query.
 
@@ -23,0 +23,0 @@ ```python

src/tasks/summarization/data.ts

 import type { TaskDataCustom } from "..";
 
 const taskData: TaskDataCustom = {
+	canonicalId: "text2text-generation",
 	datasets: [
@@ -5,0 +6,0 @@ {

src/tasks/tabular-classification/about.md

@@ -22,3 +22,3 @@ ## About the Task
 
-- Easily use inference API,
+- Easily use Inference Endpoints
 - Build neat UIs with one line of code,
@@ -25,0 +25,0 @@ - Programmatically create model cards,

src/tasks/tabular-regression/about.md

@@ -33,3 +33,3 @@ ## About the Task
 
-- Easily use inference API,
+- Easily use Inference Endpoints,
 - Build neat UIs with one line of code,
@@ -36,0 +36,0 @@ - Programmatically create model cards,

src/tasks/text-classification/about.md

@@ -153,2 +153,3 @@ ## Use Cases
 
+- [SetFitABSA: Few-Shot Aspect Based Sentiment Analysis using SetFit](https://huggingface.co/blog/setfit-absa)
 - [Course Chapter on Fine-tuning a Text Classification Model](https://huggingface.co/course/chapter3/1?fw=pt)
@@ -155,0 +156,0 @@ - [Getting Started with Sentiment Analysis using Python](https://huggingface.co/blog/sentiment-analysis-python)

src/tasks/text-generation/about.md

@@ -113,22 +113,33 @@ This task covers guides on both [text-generation](https://huggingface.co/models?pipeline_tag=text-generation&sort=downloads) and [text-to-text generation](https://huggingface.co/models?pipeline_tag=text2text-generation&sort=downloads) models. Popular large language models that are used for chats or following instructions are also covered in this task. You can find the list of selected open-source large language models [here](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard), ranked by their performance scores.
 - [Text generation strategies](https://huggingface.co/docs/transformers/generation_strategies)
+- [Course chapter on training a causal language model from scratch](https://huggingface.co/course/chapter7/6?fw=pt)
 
-### Course and Blogs
+### Model Inference & Deployment
 
-- [Course Chapter on Training a causal language model from scratch](https://huggingface.co/course/chapter7/6?fw=pt)
-- [TO Discussion with Victor Sanh](https://www.youtube.com/watch?v=Oy49SCW_Xpw&ab_channel=HuggingFace)
-- [Hugging Face Course Workshops: Pretraining Language Models & CodeParrot](https://www.youtube.com/watch?v=ExUR7w6xe94&ab_channel=HuggingFace)
-- [Training CodeParrot 🦜 from Scratch](https://huggingface.co/blog/codeparrot)
-- [How to generate text: using different decoding methods for language generation with Transformers](https://huggingface.co/blog/how-to-generate)
+- [Optimizing your LLM in production](https://huggingface.co/blog/optimize-llm)
+- [Open-Source Text Generation & LLM Ecosystem at Hugging Face](https://huggingface.co/blog/os-llms)
+- [Introducing RWKV - An RNN with the advantages of a transformer](https://huggingface.co/blog/rwkv)
+- [Llama 2 is at Hugging Face](https://huggingface.co/blog/llama2)
 - [Guiding Text Generation with Constrained Beam Search in 🤗 Transformers](https://huggingface.co/blog/constrained-beam-search)
 - [Code generation with Hugging Face](https://huggingface.co/spaces/codeparrot/code-generation-models)
-- [🌸 Introducing The World's Largest Open Multilingual Language Model: BLOOM 🌸](https://huggingface.co/blog/bloom)
-- [The Technology Behind BLOOM Training](https://huggingface.co/blog/bloom-megatron-deepspeed)
+- [Assisted Generation: a new direction toward low-latency text generation](https://huggingface.co/blog/assisted-generation)
+- [How to generate text: using different decoding methods for language generation with Transformers](https://huggingface.co/blog/how-to-generate)
 - [Faster Text Generation with TensorFlow and XLA](https://huggingface.co/blog/tf-xla-generate)
-- [Assisted Generation: a new direction toward low-latency text generation](https://huggingface.co/blog/assisted-generation)
-- [Introducing RWKV - An RNN with the advantages of a transformer](https://huggingface.co/blog/rwkv)
+
+### Model Fine-tuning/Training
+
+- [Non-engineers guide: Train a LLaMA 2 chatbot](https://huggingface.co/blog/Llama2-for-non-engineers)
+- [Training CodeParrot 🦜 from Scratch](https://huggingface.co/blog/codeparrot)
 - [Creating a Coding Assistant with StarCoder](https://huggingface.co/blog/starchat-alpha)
-- [StarCoder: A State-of-the-Art LLM for Code](https://huggingface.co/blog/starcoder)
-- [Open-Source Text Generation & LLM Ecosystem at Hugging Face](https://huggingface.co/blog/os-llms)
-- [Llama 2 is at Hugging Face](https://huggingface.co/blog/llama2)
 
+### Advanced Concepts Explained Simply
+
+- [Mixture of Experts Explained](https://huggingface.co/blog/moe)
+
+### Advanced Fine-tuning/Training Recipes
+
+- [Fine-tuning Llama 2 70B using PyTorch FSDP](https://huggingface.co/blog/ram-efficient-pytorch-fsdp)
+- [The N Implementation Details of RLHF with PPO](https://huggingface.co/blog/the_n_implementation_details_of_rlhf_with_ppo)
+- [Preference Tuning LLMs with Direct Preference Optimization Methods](https://huggingface.co/blog/pref-tuning)
+- [Fine-tune Llama 2 with DPO](https://huggingface.co/blog/dpo-trl)
+
 ### Notebooks
@@ -135,0 +146,0 @@

src/tasks/text-generation/data.ts

@@ -15,8 +15,8 @@ import type { TaskDataCustom } from "..";
 		{
-			description: "A crowd-sourced instruction dataset to develop an AI assistant.",
-			id: "OpenAssistant/oasst1",
+			description: "Truly open-source, curated and cleaned dialogue dataset.",
+			id: "HuggingFaceH4/ultrachat_200k",
 		},
 		{
-			description: "A crowd-sourced instruction dataset created by Databricks employees.",
-			id: "databricks/databricks-dolly-15k",
+			description: "An instruction dataset with preference ratings on responses.",
+			id: "openbmb/UltraFeedback",
 		},
@@ -63,31 +63,19 @@ ],
 		{
-			description: "A model trained to follow instructions, uses Pythia-12b as base model.",
-			id: "databricks/dolly-v2-12b",
+			description: "A very powerful text generation model.",
+			id: "mistralai/Mixtral-8x7B-Instruct-v0.1",
 		},
 		{
-			description: "A model trained to follow instructions curated by community, uses Pythia-12b as base model.",
-			id: "OpenAssistant/oasst-sft-4-pythia-12b-epoch-3.5",
+			description: "Small yet powerful text generation model.",
+			id: "microsoft/phi-2",
 		},
 		{
-			description: "A large language model trained to generate text in English.",
-			id: "stabilityai/stablelm-tuned-alpha-7b",
+			description: "A very powerful model that can chat, do mathematical reasoning and write code.",
+			id: "openchat/openchat-3.5-0106",
 		},
 		{
-			description: "A model trained to follow instructions, based on mosaicml/mpt-7b.",
-			id: "mosaicml/mpt-7b-instruct",
+			description: "Very strong yet small assistant model.",
+			id: "HuggingFaceH4/zephyr-7b-beta",
 		},
 		{
-			description: "A large language model trained to generate text in English.",
-			id: "EleutherAI/pythia-12b",
-		},
-		{
-			description: "A large text-to-text model trained to follow instructions.",
-			id: "google/flan-ul2",
-		},
-		{
-			description: "A large and powerful text generation model.",
-			id: "tiiuae/falcon-40b",
-		},
-		{
-			description: "State-of-the-art open-source large language model.",
+			description: "Very strong open-source large language model.",
 			id: "meta-llama/Llama-2-70b-hf",
@@ -98,12 +86,12 @@ },
 		{
-			description: "A robust text generation model that can perform various tasks through natural language prompting.",
-			id: "huggingface/bloom_demo",
+			description: "A leaderboard to compare different open-source text generation models based on various benchmarks.",
+			id: "HuggingFaceH4/open_llm_leaderboard",
 		},
 		{
-			description: "An text generation based application that can write code for 80+ languages.",
-			id: "bigcode/bigcode-playground",
+			description: "An text generation based application based on a very powerful LLaMA2 model.",
+			id: "ysharma/Explore_llamav2_with_TGI",
 		},
 		{
-			description: "An text generation based application for conversations.",
-			id: "h2oai/h2ogpt-chatbot",
+			description: "An text generation based application to converse with Zephyr model.",
+			id: "HuggingFaceH4/zephyr-chat",
 		},
@@ -115,12 +103,8 @@ {
 		{
-			description: "An text generation application that uses StableLM-tuned-alpha-7b.",
-			id: "stabilityai/stablelm-tuned-alpha-chat",
+			description: "An chatbot to converse with a very powerful text generation model.",
+			id: "mlabonne/phixtral-chat",
 		},
-		{
-			description: "An UI that uses StableLM-tuned-alpha-7b.",
-			id: "togethercomputer/OpenChatKit",
-		},
 	],
 	summary:
-		"Generating text is the task of producing new text. These models can, for example, fill in incomplete text or paraphrase.",
+		"Generating text is the task of generating new text given another text. These models can, for example, fill in incomplete text or paraphrase.",
 	widgetModels: ["HuggingFaceH4/zephyr-7b-beta"],
@@ -127,0 +111,0 @@ youtubeId: "Vpjb1lu0MDk",

src/tasks/text-to-image/about.md

@@ -56,7 +56,7 @@ ## Use Cases
 
+### Model Inference
+
 - [Hugging Face Diffusion Models Course](https://github.com/huggingface/diffusion-models-class)
 - [Getting Started with Diffusers](https://huggingface.co/docs/diffusers/index)
 - [Text-to-Image Generation](https://huggingface.co/docs/diffusers/using-diffusers/conditional_image_generation)
-- [MinImagen - Build Your Own Imagen Text-to-Image Model](https://www.assemblyai.com/blog/minimagen-build-your-own-imagen-text-to-image-model/)
-- [Using LoRA for Efficient Stable Diffusion Fine-Tuning](https://huggingface.co/blog/lora)
 - [Using Stable Diffusion with Core ML on Apple Silicon](https://huggingface.co/blog/diffusers-coreml)
@@ -66,3 +66,12 @@ - [A guide on Vector Quantized Diffusion](https://huggingface.co/blog/vq-diffusion)
 - [Running IF with 🧨 diffusers on a Free Tier Google Colab](https://huggingface.co/blog/if)
+- [Introducing Würstchen: Fast Diffusion for Image Generation](https://huggingface.co/blog/wuerstchen)
+- [Efficient Controllable Generation for SDXL with T2I-Adapters](https://huggingface.co/blog/t2i-sdxl-adapters)
+- [Welcome aMUSEd: Efficient Text-to-Image Generation](https://huggingface.co/blog/amused)
 
+### Model Fine-tuning
+
+- [Finetune Stable Diffusion Models with DDPO via TRL](https://huggingface.co/blog/pref-tuning)
+- [LoRA training scripts of the world, unite!](https://huggingface.co/blog/sdxl_lora_advanced_script)
+- [Using LoRA for Efficient Stable Diffusion Fine-Tuning](https://huggingface.co/blog/lora)
+
 This page was made possible thanks to the efforts of [Ishan Dutta](https://huggingface.co/ishandutta), [Enrique Elias Ubaldo](https://huggingface.co/herrius) and [Oğuz Akif](https://huggingface.co/oguzakif).

src/tasks/text-to-image/data.ts

@@ -82,4 +82,4 @@ import type { TaskDataCustom } from "..";
 		{
-			description: "A powerful text-to-image application that can generate 3D representations.",
-			id: "hysts/Shap-E",
+			description: "A gallery to explore various text-to-image models.",
+			id: "multimodalart/LoraTheExplorer",
 		},
@@ -90,2 +90,6 @@ {
 		},
+		{
+			description: "An application to generate realistic images given photos of a person and a prompt.",
+			id: "InstantX/InstantID",
+		},
 	],
@@ -92,0 +96,0 @@ summary:

src/tasks/text-to-speech/about.md

@@ -13,5 +13,5 @@ ## Use Cases
 
-## Inference API
+## Inference Endpoints
 
-The Hub contains over [1500 TTS models](https://huggingface.co/models?pipeline_tag=text-to-speech&sort=downloads) that you can use right away by trying out the widgets directly in the browser or calling the models as a service using the Inference API. Here is a simple code snippet to get you started:
+The Hub contains over [1500 TTS models](https://huggingface.co/models?pipeline_tag=text-to-speech&sort=downloads) that you can use right away by trying out the widgets directly in the browser or calling the models as a service using Inference Endpoints. Here is a simple code snippet to get you started:
 
@@ -65,1 +65,3 @@ ```python
 - [Speech Synthesis, Recognition, and More With SpeechT5](https://huggingface.co/blog/speecht5)
+- [Optimizing a Text-To-Speech model using 🤗 Transformers](https://huggingface.co/blog/optimizing-bark)
+-

src/tasks/text-to-speech/data.ts

 import type { TaskDataCustom } from "..";
 
 const taskData: TaskDataCustom = {
+	canonicalId: "text-to-audio",
 	datasets: [
@@ -5,0 +6,0 @@ {

src/tasks/translation/data.ts

 import type { TaskDataCustom } from "..";
 
 const taskData: TaskDataCustom = {
+	canonicalId: "text2text-generation",
 	datasets: [
@@ -5,0 +6,0 @@ {

src/tasks/zero-shot-image-classification/data.ts

@@ -55,5 +55,4 @@ import type { TaskDataCustom } from "..";
 		{
-			description:
-				"Robust image classification model trained on publicly available image-caption data trained on additional high pixel data for better performance.",
-			id: "openai/clip-vit-large-patch14-336",
+			description: "Strong zero-shot image classification model.",
+			id: "google/siglip-base-patch16-224",
 		},
@@ -68,8 +67,12 @@ {
 			description:
-				"An application that leverages zero shot image classification to find best captions to generate an image. ",
+				"An application that leverages zero-shot image classification to find best captions to generate an image. ",
 			id: "pharma/CLIP-Interrogator",
 		},
+		{
+			description: "An application to compare different zero-shot image classification models. ",
+			id: "merve/compare_clip_siglip",
+		},
 	],
 	summary:
-		"Zero shot image classification is the task of classifying previously unseen classes during training of a model.",
+		"Zero-shot image classification is the task of classifying previously unseen classes during training of a model.",
 	widgetModels: ["openai/clip-vit-large-patch14-336"],
@@ -76,0 +79,0 @@ youtubeId: "",

src/tasks/zero-shot-object-detection/about.md

 ## Use Cases
 
+Zero-shot object detection models can be used in any object detection application where the detection involves text queries for objects of interest.
+
 ### Object Search
@@ -11,2 +13,6 @@
 
+### Object Tracking
+
+Zero-shot object detectors can track objects in videos.
+
 ## Inference
@@ -13,0 +19,0 @@

src/tasks/zero-shot-object-detection/data.ts

@@ -50,3 +50,8 @@ import type { TaskDataCustom } from "..";
 	],
-	spaces: [],
+	spaces: [
+		{
+			description: "A demo to try the state-of-the-art zero-shot object detection model, OWLv2.",
+			id: "merve/owlv2",
+		},
+	],
 	summary:
@@ -53,0 +58,0 @@ "Zero-shot object detection is a computer vision task to detect objects and their classes in images, without any prior training or knowledge of the classes. Zero-shot object detection models receive an image as input, as well as a list of candidate classes, and output the bounding boxes and labels where the objects have been detected.",

tsconfig.json

@@ -5,5 +5,5 @@ {
 		"lib": ["ES2022", "DOM"],
-		"module": "CommonJS",
+		"module": "ESNext",
+		"target": "ESNext",
 		"moduleResolution": "node",
-		"target": "ES2022",
 		"forceConsistentCasingInFileNames": true,
@@ -17,4 +17,4 @@ "strict": true,
 	},
-	"include": ["src"],
+	"include": ["src", "scripts"],
 	"exclude": ["dist"]
 }

No alert changes

Improved metrics

Total package byte prevSize

901857

increased by15.83%

Number of package files

178

increased by95.6%

Lines of code

21192

increased by24.55%

Is type module

Yes

Worsened metrics

Dev dependency count

2

increased byInfinity%

No dependency changes