@huggingface/inference 1.4.3 → 1.5.0

package/LICENSE

@@ -1,6 +1,6 @@
 MIT License
 
-Copyright (c) 2022 TimMikeladze
+Copyright (c) 2022 Tim Mikeladze
 
 Permission is hereby granted, free of charge, to any person obtaining a copy
 of this software and associated documentation files (the "Software"), to deal
@@ -18,4 +18,4 @@ FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
 AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
 LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
 OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
-SOFTWARE.
+SOFTWARE.

package/README.md

@@ -137,6 +137,12 @@ await hf.imageSegmentation({
   data: readFileSync('test/cats.png'),
   model: 'facebook/detr-resnet-50-panoptic'
 })
+
+await hf.textToImage({
+  inputs: 'award winning high resolution photo of a giant tortoise/((ladybird)) hybrid, [trending on artstation]',
+  negative_prompt: 'blurry',
+  model: 'stabilityai/stable-diffusion-2',
+})
 ```
 
 ## Supported APIs
@@ -167,6 +173,7 @@ await hf.imageSegmentation({
 - [x] Image classification
 - [x] Object detection
 - [x] Image segmentation
+- [x] Text to image
 
 ## Running tests
 
@@ -296,6 +303,12 @@ export declare class HfInference {
     args: ImageSegmentationArgs,
     options?: Options
   ): Promise<ImageSegmentationReturn>
+
+  /**
+   * This task reads some text input and outputs an image.
+   * Recommended model: stabilityai/stable-diffusion-2
+   */
+  textToImage(args: TextToImageArgs, options?: Options): Promise<TextToImageReturn>;
   request(
     args: Args & {
       data?: any
@@ -752,4 +765,15 @@ export declare type AudioClassificationReturnValue = {
   score: number
 }
 export declare type AudioClassificationReturn = AudioClassificationReturnValue[]
+type TextToImageArgs = Args & {
+  /**
+   * The text to generate an image from
+   */
+  inputs: string
+  /**
+   * An optional negative prompt for the image generation
+   */
+  negative_prompt?: string
+};
+type TextToImageReturn = ArrayBuffer
 ```

package/dist/index.d.ts

@@ -442,6 +442,17 @@ type AudioClassificationReturnValue = {
     score: number;
 };
 type AudioClassificationReturn = AudioClassificationReturnValue[];
+type TextToImageArgs = Args & {
+    /**
+     * The text to generate an image from
+     */
+    inputs: string;
+    /**
+     * An optional negative prompt for the image generation
+     */
+    negative_prompt?: string;
+};
+type TextToImageReturn = ArrayBuffer;
 declare class HfInference {
     private readonly apiKey;
     private readonly defaultOptions;
@@ -516,12 +527,18 @@ declare class HfInference {
      * Recommended model: facebook/detr-resnet-50-panoptic
      */
     imageSegmentation(args: ImageSegmentationArgs, options?: Options): Promise<ImageSegmentationReturn>;
+    /**
+     * This task reads some text input and outputs an image.
+     * Recommended model: stabilityai/stable-diffusion-2
+     */
+    textToImage(args: TextToImageArgs, options?: Options): Promise<TextToImageReturn>;
     request(args: Args & {
         data?: any;
     }, options?: Options & {
         binary?: boolean;
+        blob?: boolean;
     }): Promise<any>;
     private static toArray;
 }
 
-export { Args, AudioClassificationArgs, AudioClassificationReturn, AudioClassificationReturnValue, AutomaticSpeechRecognitionArgs, AutomaticSpeechRecognitionReturn, ConversationalArgs, ConversationalReturn, FeatureExtractionArgs, FeatureExtractionReturn, FillMaskArgs, FillMaskReturn, HfInference, ImageClassificationArgs, ImageClassificationReturn, ImageClassificationReturnValue, ImageSegmentationArgs, ImageSegmentationReturn, ImageSegmentationReturnValue, ObjectDetectionArgs, ObjectDetectionReturn, ObjectDetectionReturnValue, Options, QuestionAnswerArgs, QuestionAnswerReturn, SummarizationArgs, SummarizationReturn, TableQuestionAnswerArgs, TableQuestionAnswerReturn, TextClassificationArgs, TextClassificationReturn, TextGenerationArgs, TextGenerationReturn, TokenClassificationArgs, TokenClassificationReturn, TokenClassificationReturnValue, TranslationArgs, TranslationReturn, ZeroShotClassificationArgs, ZeroShotClassificationReturn, ZeroShotClassificationReturnValue };
+export { Args, AudioClassificationArgs, AudioClassificationReturn, AudioClassificationReturnValue, AutomaticSpeechRecognitionArgs, AutomaticSpeechRecognitionReturn, ConversationalArgs, ConversationalReturn, FeatureExtractionArgs, FeatureExtractionReturn, FillMaskArgs, FillMaskReturn, HfInference, ImageClassificationArgs, ImageClassificationReturn, ImageClassificationReturnValue, ImageSegmentationArgs, ImageSegmentationReturn, ImageSegmentationReturnValue, ObjectDetectionArgs, ObjectDetectionReturn, ObjectDetectionReturnValue, Options, QuestionAnswerArgs, QuestionAnswerReturn, SummarizationArgs, SummarizationReturn, TableQuestionAnswerArgs, TableQuestionAnswerReturn, TextClassificationArgs, TextClassificationReturn, TextGenerationArgs, TextGenerationReturn, TextToImageArgs, TextToImageReturn, TokenClassificationArgs, TokenClassificationReturn, TokenClassificationReturnValue, TranslationArgs, TranslationReturn, ZeroShotClassificationArgs, ZeroShotClassificationReturn, ZeroShotClassificationReturnValue };

package/dist/index.js

@@ -148,11 +148,27 @@ var HfInference = class {
       binary: true
     });
   }
+  /**
+   * This task reads some text input and outputs an image.
+   * Recommended model: stabilityai/stable-diffusion-2
+   */
+  async textToImage(args, options) {
+    return await this.request(args, {
+      ...options,
+      blob: true
+    });
+  }
   async request(args, options) {
     const mergedOptions = { ...this.defaultOptions, ...options };
     const { model, ...otherArgs } = args;
+    const headers = {
+      Authorization: `Bearer ${this.apiKey}`
+    };
+    if (options?.binary && mergedOptions.wait_for_model) {
+      headers["X-Wait-For-Model"] = "true";
+    }
     const response = await fetch(`https://api-inference.huggingface.co/models/${model}`, {
-      headers: { Authorization: `Bearer ${this.apiKey}` },
+      headers,
       method: "POST",
       body: options?.binary ? args.data : JSON.stringify({
         ...otherArgs,
@@ -165,11 +181,19 @@ var HfInference = class {
         wait_for_model: true
       });
     }
-    const res = await response.json();
-    if (res.error) {
-      throw new Error(res.error);
+    let output;
+    if (options?.blob) {
+      if (!response.ok) {
+        throw new Error("An error occurred while fetching the blob");
+      }
+      return await response.arrayBuffer();
+    } else {
+      output = await response.json();
+      if (output.error) {
+        throw new Error(output.error);
+      }
     }
-    return res;
+    return output;
   }
   static toArray(obj) {
     if (Array.isArray(obj)) {

package/dist/index.mjs

@@ -123,11 +123,27 @@ var HfInference = class {
       binary: true
     });
   }
+  /**
+   * This task reads some text input and outputs an image.
+   * Recommended model: stabilityai/stable-diffusion-2
+   */
+  async textToImage(args, options) {
+    return await this.request(args, {
+      ...options,
+      blob: true
+    });
+  }
   async request(args, options) {
     const mergedOptions = { ...this.defaultOptions, ...options };
     const { model, ...otherArgs } = args;
+    const headers = {
+      Authorization: `Bearer ${this.apiKey}`
+    };
+    if (options?.binary && mergedOptions.wait_for_model) {
+      headers["X-Wait-For-Model"] = "true";
+    }
     const response = await fetch(`https://api-inference.huggingface.co/models/${model}`, {
-      headers: { Authorization: `Bearer ${this.apiKey}` },
+      headers,
       method: "POST",
       body: options?.binary ? args.data : JSON.stringify({
         ...otherArgs,
@@ -140,11 +156,19 @@ var HfInference = class {
         wait_for_model: true
       });
     }
-    const res = await response.json();
-    if (res.error) {
-      throw new Error(res.error);
+    let output;
+    if (options?.blob) {
+      if (!response.ok) {
+        throw new Error("An error occurred while fetching the blob");
+      }
+      return await response.arrayBuffer();
+    } else {
+      output = await response.json();
+      if (output.error) {
+        throw new Error(output.error);
+      }
     }
-    return res;
+    return output;
   }
   static toArray(obj) {
     if (Array.isArray(obj)) {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@huggingface/inference",
-  "version": "1.4.3",
+  "version": "1.5.0",
   "license": "MIT",
   "author": "Tim Mikeladze <tim.mikeladze@gmail.com>",
   "description": "Typescript wrapper for the Hugging Face Inference API",

package/src/HfInference.ts

@@ -485,6 +485,20 @@ export type AudioClassificationReturnValue = {
 
 export type AudioClassificationReturn = AudioClassificationReturnValue[];
 
+export type TextToImageArgs = Args & {
+	/**
+	 * The text to generate an image from
+	 */
+	inputs: string;
+
+	/**
+	 * An optional negative prompt for the image generation
+	 */
+	negative_prompt?: string;
+};
+
+export type TextToImageReturn = ArrayBuffer;
+
 export class HfInference {
 	private readonly apiKey:         string;
 	private readonly defaultOptions: Options;
@@ -645,18 +659,39 @@ export class HfInference {
 		});
 	}
 
+	/**
+	 * This task reads some text input and outputs an image.
+	 * Recommended model: stabilityai/stable-diffusion-2
+	 */
+	public async textToImage(args: TextToImageArgs, options?: Options): Promise<TextToImageReturn> {
+		return await this.request(args, {
+			...options,
+			blob: true,
+		});
+	}
+
 	public async request(
 		args: Args & { data?: any },
 		options?: Options & {
 			binary?: boolean;
+			blob?:   boolean;
 		}
 	): Promise<any> {
 		const mergedOptions = { ...this.defaultOptions, ...options };
 		const { model, ...otherArgs } = args;
+
+    const headers = {
+      Authorization: `Bearer ${this.apiKey}`,
+    }
+
+    if (options?.binary && mergedOptions.wait_for_model) {
+      headers["X-Wait-For-Model"] = "true";
+    }
+
 		const response = await fetch(`https://api-inference.huggingface.co/models/${model}`, {
-			headers: { Authorization: `Bearer ${this.apiKey}` },
-			method:  "POST",
-			body:    options?.binary
+			headers,
+			method: "POST",
+			body:   options?.binary
 				? args.data
 				: JSON.stringify({
 						...otherArgs,
@@ -671,11 +706,21 @@ export class HfInference {
 			});
 		}
 
-		const res = await response.json();
-		if (res.error) {
-			throw new Error(res.error);
+		let output;
+
+		if (options?.blob) {
+			if (!response.ok) {
+				throw new Error("An error occurred while fetching the blob");
+			}
+			return await response.arrayBuffer();
+		} else {
+			output = await response.json();
+			if (output.error) {
+				throw new Error(output.error);
+			}
 		}
-		return res;
+
+		return output;
 	}
 
 	private static toArray(obj: any): any[] {