@huggingface/inference 1.7.1 → 2.0.0-rc1

package/README.md

@@ -1,10 +1,10 @@
 # 🤗 Hugging Face Inference API
 
-A Typescript powered wrapper for the Hugging Face Inference API. Learn more about the Inference API at [Hugging Face](https://huggingface.co/docs/api-inference/index).
-
-Check out the [full documentation](https://huggingface.co/docs/huggingface.js/inference/README) or try out a live [interactive notebook](https://observablehq.com/@huggingface/hello-huggingface-js-inference).
+A Typescript powered wrapper for the Hugging Face Inference API. Learn more about the Inference API at [Hugging Face](https://huggingface.co/docs/api-inference/index). It also works with [Inference Endpoints](https://huggingface.co/docs/inference-endpoints/index).
 
+Check out the [full documentation](https://huggingface.co/docs/huggingface.js/inference/README).
 
+You can also try out a live [interactive notebook](https://observablehq.com/@huggingface/hello-huggingface-js-inference) or see some demos on [hf.co/huggingfacejs](https://huggingface.co/huggingfacejs).
 
 ## Install
 
@@ -18,16 +18,16 @@ pnpm add @huggingface/inference
 
 ## Usage
 
-❗**Important note:** Using an API key is optional to get started, however you will be rate limited eventually. Join [Hugging Face](https://huggingface.co/join) and then visit [access tokens](https://huggingface.co/settings/tokens) to generate your API key for **free**. 
+❗**Important note:** Using an access token is optional to get started, however you will be rate limited eventually. Join [Hugging Face](https://huggingface.co/join) and then visit [access tokens](https://huggingface.co/settings/tokens) to generate your access token for **free**.
 
-Your API key should be kept private. If you need to protect it in front-end applications, we suggest setting up a proxy server that stores the API key.
+Your access token should be kept private. If you need to protect it in front-end applications, we suggest setting up a proxy server that stores the access token.
 
 ### Basic examples
 
 ```typescript
 import { HfInference } from '@huggingface/inference'
 
-const hf = new HfInference('your api key')
+const hf = new HfInference('your access token')
 
 // Natural Language
 
@@ -45,7 +45,7 @@ await hf.summarization({
   }
 })
 
-await hf.questionAnswer({
+await hf.questionAnswering({
   model: 'deepset/roberta-base-squad2',
   inputs: {
     question: 'What is the capital of France?',
@@ -53,7 +53,7 @@ await hf.questionAnswer({
   }
 })
 
-await hf.tableQuestionAnswer({
+await hf.tableQuestionAnswering({
   model: 'google/tapas-base-finetuned-wtq',
   inputs: {
     query: 'How many stars does the transformers repository have?',
@@ -111,7 +111,7 @@ await hf.conversational({
   }
 })
 
-await hf.featureExtraction({
+await hf.sentenceSimilarity({
   model: 'sentence-transformers/paraphrase-xlm-r-multilingual-v1',
   inputs: {
     source_sentence: 'That is a happy person',
@@ -123,6 +123,11 @@ await hf.featureExtraction({
   }
 })
 
+await hf.featureExtraction({
+    model: "sentence-transformers/distilbert-base-nli-mean-tokens",
+    inputs: "That is a happy person",
+});
+
 // Audio
 
 await hf.automaticSpeechRecognition({
@@ -154,9 +159,40 @@ await hf.imageSegmentation({
 
 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',
+  parameters: {
+    negative_prompt: 'blurry',
+  }
+})
+
+await hf.imageToText({
+  data: readFileSync('test/cats.png'),
+  model: 'nlpconnect/vit-gpt2-image-captioning'
+})
+
+// Custom call, for models with custom parameters / outputs
+await hf.request({
+  model: 'my-custom-model',
+  inputs: 'hello world',
+  parameters: {
+    custom_param: 'some magic',
+  }
 })
+
+// Custom streaming call, for models with custom parameters / outputs
+for await (const output of hf.streamingRequest({
+  model: 'my-custom-model',
+  inputs: 'hello world',
+  parameters: {
+    custom_param: 'some magic',
+  }
+})) {
+  ...
+}
+
+// Using your own inference endpoint: https://hf.co/docs/inference-endpoints/
+const gpt2 = hf.endpoint('https://xyz.eu-west-1.aws.endpoints.huggingface.cloud/gpt2');
+const { generated_text } = await gpt2.textGeneration({inputs: 'The answer to the universe is'});
 ```
 
 ## Supported Tasks
@@ -176,6 +212,7 @@ await hf.textToImage({
 - [x] Zero-shot classification
 - [x] Conversational
 - [x] Feature extraction
+- [x] Sentence Similarity
 
 ### Audio
 
@@ -188,6 +225,24 @@ await hf.textToImage({
 - [x] Object detection
 - [x] Image segmentation
 - [x] Text to image
+- [x] Image to text
+
+## Tree-shaking
+
+You can import the functions you need directly from the module, rather than using the `HfInference` class:
+
+```ts
+import {textGeneration} from "@huggingface/inference";
+
+await textGeneration({
+  accessToken: "hf_...",
+  model: "model_or_endpoint",
+  inputs: ...,
+  parameters: ...
+})
+```
+
+This will enable tree-shaking by your bundler.
 
 ## Running tests