🔖 @hugginface/inference v1.7.0

Author: machineuser

README.md

@@ -48,7 +48,7 @@ You can run our packages with vanilla JS, without any bundler, by using a CDN or
 ```html
 
 <script type="module">
-    import { HfInference } from 'https://cdn.jsdelivr.net/npm/@huggingface/inference@1.6.3/+esm';
+    import { HfInference } from 'https://cdn.jsdelivr.net/npm/@huggingface/inference@1.7.0/+esm';
     import { createRepo, commit, deleteRepo, listFiles } from "https://cdn.jsdelivr.net/npm/@huggingface/[email protected]/+esm";
 </script>
 ```

docs/_toctree.yml

@@ -62,6 +62,10 @@
       sections:
         - title: HfInference
           local: inference/classes/HfInference
+    - title: Enums
+      sections:
+        - title: TextGenerationStreamFinishReason
+          local: inference/enums/TextGenerationStreamFinishReason
     - title: Interfaces
       sections:
         - title: Args
@@ -88,6 +92,16 @@
           local: inference/interfaces/TableQuestionAnswerReturn
         - title: TextGenerationReturn
           local: inference/interfaces/TextGenerationReturn
+        - title: TextGenerationStreamBestOfSequence
+          local: inference/interfaces/TextGenerationStreamBestOfSequence
+        - title: TextGenerationStreamDetails
+          local: inference/interfaces/TextGenerationStreamDetails
+        - title: TextGenerationStreamPrefillToken
+          local: inference/interfaces/TextGenerationStreamPrefillToken
+        - title: TextGenerationStreamReturn
+          local: inference/interfaces/TextGenerationStreamReturn
+        - title: TextGenerationStreamToken
+          local: inference/interfaces/TextGenerationStreamToken
         - title: TokenClassificationReturnValue
           local: inference/interfaces/TokenClassificationReturnValue
         - title: TranslationReturn

docs/index.md

@@ -48,7 +48,7 @@ You can run our packages with vanilla JS, without any bundler, by using a CDN or
 ```html
 
 <script type="module">
-    import { HfInference } from 'https://cdn.jsdelivr.net/npm/@huggingface/inference@1.6.3/+esm';
+    import { HfInference } from 'https://cdn.jsdelivr.net/npm/@huggingface/inference@1.7.0/+esm';
     import { createRepo, commit, deleteRepo, listFiles } from "https://cdn.jsdelivr.net/npm/@huggingface/[email protected]/+esm";
 </script>
 ```

docs/inference/README.md

@@ -72,6 +72,13 @@ await hf.textGeneration({
   inputs: 'The answer to the universe is'
 })
 
+for await const (output of hf.textGenerationStream({
+  model: "google/flan-t5-xxl",
+  inputs: 'repeat "one two three four"'
+})) {
+  console.log(output.token.text, output.generated_text);
+}
+
 await hf.tokenClassification({
   model: 'dbmdz/bert-large-cased-finetuned-conll03-english',
   inputs: 'My name is Sarah Jessica Parker but you can call me Jessica'

docs/inference/classes/HfInference.md

@@ -15,7 +15,7 @@
 
 #### Defined in
 
-[HfInference.ts:513](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L513)
+[HfInference.ts:597](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L597)
 
 ## Properties
 
@@ -25,7 +25,7 @@
 
 #### Defined in
 
-[HfInference.ts:510](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L510)
+[HfInference.ts:594](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L594)
 
 ___
 
@@ -35,7 +35,7 @@ ___
 
 #### Defined in
 
-[HfInference.ts:511](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L511)
+[HfInference.ts:595](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L595)
 
 ## Methods
 
@@ -59,7 +59,7 @@ Recommended model:  superb/hubert-large-superb-er
 
 #### Defined in
 
-[HfInference.ts:737](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L737)
+[HfInference.ts:831](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L831)
 
 ___
 
@@ -83,7 +83,7 @@ Recommended model (english language): facebook/wav2vec2-large-960h-lv60-self
 
 #### Defined in
 
-[HfInference.ts:718](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L718)
+[HfInference.ts:812](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L812)
 
 ___
 
@@ -106,7 +106,7 @@ This task corresponds to any chatbot like structure. Models tend to have shorter
 
 #### Defined in
 
-[HfInference.ts:688](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L688)
+[HfInference.ts:782](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L782)
 
 ___
 
@@ -129,7 +129,7 @@ This task reads some text and outputs raw float values, that are usually consume
 
 #### Defined in
 
-[HfInference.ts:709](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L709)
+[HfInference.ts:803](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L803)
 
 ___
 
@@ -152,7 +152,7 @@ Tries to fill in a hole with a missing word (token to be precise). That’s the
 
 #### Defined in
 
-[HfInference.ts:521](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L521)
+[HfInference.ts:605](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L605)
 
 ___
 
@@ -176,7 +176,7 @@ Recommended model: google/vit-base-patch16-224
 
 #### Defined in
 
-[HfInference.ts:757](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L757)
+[HfInference.ts:851](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L851)
 
 ___
 
@@ -200,7 +200,44 @@ Recommended model: facebook/detr-resnet-50-panoptic
 
 #### Defined in
 
-[HfInference.ts:805](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L805)
+[HfInference.ts:899](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L899)
+
+___
+
+### makeRequestOptions
+
+▸ `Private` **makeRequestOptions**(`args`, `options?`): `Object`
+
+Helper that prepares request arguments
+
+#### Parameters
+
+| Name | Type |
+| :------ | :------ |
+| `args` | [`Args`](../interfaces/Args) & { `data?`: `ArrayBuffer` \| `Blob` ; `stream?`: `boolean`  } |
+| `options?` | [`Options`](../interfaces/Options) & { `binary?`: `boolean` ; `blob?`: `boolean` ; `includeCredentials?`: `boolean`  } |
+
+#### Returns
+
+`Object`
+
+| Name | Type |
+| :------ | :------ |
+| `info` | `RequestInit` |
+| `mergedOptions` | { `binary?`: `boolean` ; `blob?`: `boolean` ; `dont_load_model?`: `boolean` ; `includeCredentials?`: `boolean` ; `retry_on_error?`: `boolean` ; `use_cache?`: `boolean` ; `use_gpu?`: `boolean` ; `wait_for_model?`: `boolean`  } |
+| `mergedOptions.binary?` | `boolean` |
+| `mergedOptions.blob?` | `boolean` |
+| `mergedOptions.dont_load_model?` | `boolean` |
+| `mergedOptions.includeCredentials?` | `boolean` |
+| `mergedOptions.retry_on_error?` | `boolean` |
+| `mergedOptions.use_cache?` | `boolean` |
+| `mergedOptions.use_gpu?` | `boolean` |
+| `mergedOptions.wait_for_model?` | `boolean` |
+| `url` | `string` |
+
+#### Defined in
+
+[HfInference.ts:934](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L934)
 
 ___
 
@@ -224,7 +261,7 @@ Recommended model: facebook/detr-resnet-50
 
 #### Defined in
 
-[HfInference.ts:777](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L777)
+[HfInference.ts:871](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L871)
 
 ___
 
@@ -247,7 +284,7 @@ Want to have a nice know-it-all bot that can answer any question?. Recommended m
 
 #### Defined in
 
-[HfInference.ts:555](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L555)
+[HfInference.ts:639](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L639)
 
 ___
 
@@ -274,7 +311,36 @@ ___
 
 #### Defined in
 
-[HfInference.ts:837](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L837)
+[HfInference.ts:986](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L986)
+
+___
+
+### streamingRequest
+
+▸ **streamingRequest**<`T`\>(`args`, `options?`): `AsyncGenerator`<`T`, `any`, `unknown`\>
+
+Make request that uses server-sent events and returns response as a generator
+
+#### Type parameters
+
+| Name |
+| :------ |
+| `T` |
+
+#### Parameters
+
+| Name | Type |
+| :------ | :------ |
+| `args` | [`Args`](../interfaces/Args) & { `data?`: `ArrayBuffer` \| `Blob`  } |
+| `options?` | [`Options`](../interfaces/Options) & { `binary?`: `boolean` ; `blob?`: `boolean` ; `includeCredentials?`: `boolean`  } |
+
+#### Returns
+
+`AsyncGenerator`<`T`, `any`, `unknown`\>
+
+#### Defined in
+
+[HfInference.ts:1022](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L1022)
 
 ___
 
@@ -297,7 +363,7 @@ This task is well known to summarize longer text into shorter text. Be careful,
 
 #### Defined in
 
-[HfInference.ts:543](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L543)
+[HfInference.ts:627](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L627)
 
 ___
 
@@ -320,7 +386,7 @@ Don’t know SQL? Don’t want to dive into a large spreadsheet? Ask questions i
 
 #### Defined in
 
-[HfInference.ts:573](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L573)
+[HfInference.ts:657](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L657)
 
 ___
 
@@ -343,7 +409,7 @@ Usually used for sentiment-analysis this will output the likelihood of classes o
 
 #### Defined in
 
-[HfInference.ts:596](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L596)
+[HfInference.ts:680](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L680)
 
 ___
 
@@ -366,7 +432,30 @@ Use to continue text from a prompt. This is a very generic task. Recommended mod
 
 #### Defined in
 
-[HfInference.ts:609](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L609)
+[HfInference.ts:693](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L693)
+
+___
+
+### textGenerationStream
+
+▸ **textGenerationStream**(`args`, `options?`): `AsyncGenerator`<[`TextGenerationStreamReturn`](../interfaces/TextGenerationStreamReturn), `any`, `unknown`\>
+
+Use to continue text from a prompt. Same as `textGeneration` but returns generator that can be read one token at a time
+
+#### Parameters
+
+| Name | Type |
+| :------ | :------ |
+| `args` | [`TextGenerationArgs`](../modules#textgenerationargs) |
+| `options?` | [`Options`](../interfaces/Options) |
+
+#### Returns
+
+`AsyncGenerator`<[`TextGenerationStreamReturn`](../interfaces/TextGenerationStreamReturn), `any`, `unknown`\>
+
+#### Defined in
+
+[HfInference.ts:705](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L705)
 
 ___
 
@@ -390,7 +479,7 @@ Recommended model: stabilityai/stable-diffusion-2
 
 #### Defined in
 
-[HfInference.ts:825](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L825)
+[HfInference.ts:919](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L919)
 
 ___
 
@@ -413,7 +502,7 @@ Usually used for sentence parsing, either grammatical, or Named Entity Recogniti
 
 #### Defined in
 
-[HfInference.ts:621](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L621)
+[HfInference.ts:715](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L715)
 
 ___
 
@@ -436,7 +525,7 @@ This task is well known to translate text from one language to another. Recommen
 
 #### Defined in
 
-[HfInference.ts:647](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L647)
+[HfInference.ts:741](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L741)
 
 ___
 
@@ -459,4 +548,4 @@ This task is super useful to try out classification with zero code, you simply p
 
 #### Defined in
 
-[HfInference.ts:659](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L659)
+[HfInference.ts:753](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L753)

docs/inference/enums/TextGenerationStreamFinishReason.md

@@ -0,0 +1,37 @@
+# Enumeration: TextGenerationStreamFinishReason
+
+## Enumeration Members
+
+### EndOfSequenceToken
+
+• **EndOfSequenceToken** = ``"eos_token"``
+
+the model generated its end of sequence token
+
+#### Defined in
+
+[HfInference.ts:275](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L275)
+
+___
+
+### Length
+
+• **Length** = ``"length"``
+
+number of generated tokens == `max_new_tokens`
+
+#### Defined in
+
+[HfInference.ts:273](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L273)
+
+___
+
+### StopSequence
+
+• **StopSequence** = ``"stop_sequence"``
+
+the model generated a text included in `stop_sequences`
+
+#### Defined in
+
+[HfInference.ts:277](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L277)

docs/inference/interfaces/Args.md

@@ -8,4 +8,4 @@
 
 #### Defined in
 
-[HfInference.ts:28](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L28)
+[HfInference.ts:32](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L32)

docs/inference/interfaces/AudioClassificationReturnValue.md

@@ -10,7 +10,7 @@ The label for the class (model specific)
 
 #### Defined in
 
-[HfInference.ts:485](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L485)
+[HfInference.ts:569](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L569)
 
 ___
 
@@ -22,4 +22,4 @@ A float that represents how likely it is that the audio file belongs to this cla
 
 #### Defined in
 
-[HfInference.ts:490](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L490)
+[HfInference.ts:574](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L574)

docs/inference/interfaces/AutomaticSpeechRecognitionReturn.md

@@ -10,4 +10,4 @@ The text that was recognized from the audio
 
 #### Defined in
 
-[HfInference.ts:471](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L471)
+[HfInference.ts:555](https://github.com/huggingface/huggingface.js/blob/main/packages/inference/src/HfInference.ts#L555)