Merge pull request #2 from huggingface/main

nico-martin · web-flow · commit 19a6e0318f3d · 2025-10-10T13:25:56.000+02:00
merged latest
diff --git a/.gitignore b/.gitignore
@@ -4,6 +4,7 @@ node_modules
 .cache
 .DS_STORE
 deno.lock
+.idea
 
 # Do not track build artifacts/generated files
 /dist
diff --git a/README.md b/README.md
@@ -47,7 +47,7 @@ npm i @huggingface/transformers
 Alternatively, you can use it in vanilla JS, without any bundler, by using a CDN or static hosting. For example, using [ES Modules](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Guide/Modules), you can import the library with:
 ```html
 <script type="module">
-    import { pipeline } from 'https://cdn.jsdelivr.net/npm/@huggingface/transformers@3.7.4';
+    import { pipeline } from 'https://cdn.jsdelivr.net/npm/@huggingface/transformers@3.7.5';
 </script>
 ```
 
@@ -155,7 +155,7 @@ Check out the Transformers.js [template](https://huggingface.co/new-space?templa
 
 
 
-By default, Transformers.js uses [hosted pretrained models](https://huggingface.co/models?library=transformers.js) and [precompiled WASM binaries](https://cdn.jsdelivr.net/npm/@huggingface/transformers@3.7.4/dist/), which should work out-of-the-box. You can customize this as follows:
+By default, Transformers.js uses [hosted pretrained models](https://huggingface.co/models?library=transformers.js) and [precompiled WASM binaries](https://cdn.jsdelivr.net/npm/@huggingface/transformers@3.7.5/dist/), which should work out-of-the-box. You can customize this as follows:
 
 ### Settings
 
@@ -344,6 +344,7 @@ You can refine your search by selecting the task you're interested in (e.g., [te
 1. **[GPT-J](https://huggingface.co/docs/transformers/model_doc/gptj)** (from EleutherAI) released in the repository [kingoflolz/mesh-transformer-jax](https://github.com/kingoflolz/mesh-transformer-jax/) by Ben Wang and Aran Komatsuzaki.
 1. **[GPTBigCode](https://huggingface.co/docs/transformers/model_doc/gpt_bigcode)** (from BigCode) released with the paper [SantaCoder: don't reach for the stars!](https://huggingface.co/papers/2301.03988) by Loubna Ben Allal, Raymond Li, Denis Kocetkov, Chenghao Mou, Christopher Akiki, Carlos Munoz Ferrandis, Niklas Muennighoff, Mayank Mishra, Alex Gu, Manan Dey, Logesh Kumar Umapathi, Carolyn Jane Anderson, Yangtian Zi, Joel Lamy Poirier, Hailey Schoelkopf, Sergey Troshin, Dmitry Abulkhanov, Manuel Romero, Michael Lappert, Francesco De Toni, Bernardo García del Río, Qian Liu, Shamik Bose, Urvashi Bhattacharyya, Terry Yue Zhuo, Ian Yu, Paulo Villegas, Marco Zocca, Sourab Mangrulkar, David Lansky, Huu Nguyen, Danish Contractor, Luis Villa, Jia Li, Dzmitry Bahdanau, Yacine Jernite, Sean Hughes, Daniel Fried, Arjun Guha, Harm de Vries, Leandro von Werra.
 1. **[Granite](https://huggingface.co/docs/transformers/main/model_doc/granite)** (from IBM) released with the paper [Power Scheduler: A Batch Size and Token Number Agnostic Learning Rate Scheduler](https://huggingface.co/papers/2408.13359) by Yikang Shen, Matthew Stallone, Mayank Mishra, Gaoyuan Zhang, Shawn Tan, Aditya Prasad, Adriana Meza Soria, David D. Cox, Rameswar Panda.
+1. **[GraniteMoeHybrid](https://huggingface.co/docs/transformers/main/model_doc/granitemoehybrid)** (from IBM) released with the blog post [IBM Granite 4.0: hyper-efficient, high performance hybrid models for enterprise](https://www.ibm.com/new/announcements/ibm-granite-4-0-hyper-efficient-high-performance-hybrid-models) by the IBM Granite team.
 1. **[Grounding DINO](https://huggingface.co/docs/transformers/model_doc/grounding-dino)** (from IDEA-Research) released with the paper [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://huggingface.co/papers/2303.05499) by Shilong Liu, Zhaoyang Zeng, Tianhe Ren, Feng Li, Hao Zhang, Jie Yang, Qing Jiang, Chunyuan Li, Jianwei Yang, Hang Su, Jun Zhu, Lei Zhang.
 1. **[GroupViT](https://huggingface.co/docs/transformers/model_doc/groupvit)** (from UCSD, NVIDIA) released with the paper [GroupViT: Semantic Segmentation Emerges from Text Supervision](https://huggingface.co/papers/2202.11094) by Jiarui Xu, Shalini De Mello, Sifei Liu, Wonmin Byeon, Thomas Breuel, Jan Kautz, Xiaolong Wang.
 1. **[Helium](https://huggingface.co/docs/transformers/main/model_doc/helium)** (from the Kyutai Team) released with the blog post [Announcing Helium-1 Preview](https://kyutai.org/2025/01/13/helium.html) by the Kyutai Team.
diff --git a/docs/snippets/2_installation.snippet b/docs/snippets/2_installation.snippet
@@ -7,6 +7,6 @@ npm i @huggingface/transformers
 Alternatively, you can use it in vanilla JS, without any bundler, by using a CDN or static hosting. For example, using [ES Modules](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Guide/Modules), you can import the library with:
 ```html
 <script type="module">
-    import { pipeline } from 'https://cdn.jsdelivr.net/npm/@huggingface/transformers@3.7.4';
+    import { pipeline } from 'https://cdn.jsdelivr.net/npm/@huggingface/transformers@3.7.5';
 </script>
 ```
diff --git a/docs/snippets/4_custom-usage.snippet b/docs/snippets/4_custom-usage.snippet
@@ -1,6 +1,6 @@
 
 
-By default, Transformers.js uses [hosted pretrained models](https://huggingface.co/models?library=transformers.js) and [precompiled WASM binaries](https://cdn.jsdelivr.net/npm/@huggingface/transformers@3.7.4/dist/), which should work out-of-the-box. You can customize this as follows:
+By default, Transformers.js uses [hosted pretrained models](https://huggingface.co/models?library=transformers.js) and [precompiled WASM binaries](https://cdn.jsdelivr.net/npm/@huggingface/transformers@3.7.5/dist/), which should work out-of-the-box. You can customize this as follows:
 
 ### Settings
 
diff --git a/docs/snippets/6_supported-models.snippet b/docs/snippets/6_supported-models.snippet
@@ -58,6 +58,7 @@
 1. **[GPT-J](https://huggingface.co/docs/transformers/model_doc/gptj)** (from EleutherAI) released in the repository [kingoflolz/mesh-transformer-jax](https://github.com/kingoflolz/mesh-transformer-jax/) by Ben Wang and Aran Komatsuzaki.
 1. **[GPTBigCode](https://huggingface.co/docs/transformers/model_doc/gpt_bigcode)** (from BigCode) released with the paper [SantaCoder: don't reach for the stars!](https://huggingface.co/papers/2301.03988) by Loubna Ben Allal, Raymond Li, Denis Kocetkov, Chenghao Mou, Christopher Akiki, Carlos Munoz Ferrandis, Niklas Muennighoff, Mayank Mishra, Alex Gu, Manan Dey, Logesh Kumar Umapathi, Carolyn Jane Anderson, Yangtian Zi, Joel Lamy Poirier, Hailey Schoelkopf, Sergey Troshin, Dmitry Abulkhanov, Manuel Romero, Michael Lappert, Francesco De Toni, Bernardo García del Río, Qian Liu, Shamik Bose, Urvashi Bhattacharyya, Terry Yue Zhuo, Ian Yu, Paulo Villegas, Marco Zocca, Sourab Mangrulkar, David Lansky, Huu Nguyen, Danish Contractor, Luis Villa, Jia Li, Dzmitry Bahdanau, Yacine Jernite, Sean Hughes, Daniel Fried, Arjun Guha, Harm de Vries, Leandro von Werra.
 1. **[Granite](https://huggingface.co/docs/transformers/main/model_doc/granite)** (from IBM) released with the paper [Power Scheduler: A Batch Size and Token Number Agnostic Learning Rate Scheduler](https://huggingface.co/papers/2408.13359) by Yikang Shen, Matthew Stallone, Mayank Mishra, Gaoyuan Zhang, Shawn Tan, Aditya Prasad, Adriana Meza Soria, David D. Cox, Rameswar Panda.
+1. **[GraniteMoeHybrid](https://huggingface.co/docs/transformers/main/model_doc/granitemoehybrid)** (from IBM) released with the blog post [IBM Granite 4.0: hyper-efficient, high performance hybrid models for enterprise](https://www.ibm.com/new/announcements/ibm-granite-4-0-hyper-efficient-high-performance-hybrid-models) by the IBM Granite team.
 1. **[Grounding DINO](https://huggingface.co/docs/transformers/model_doc/grounding-dino)** (from IDEA-Research) released with the paper [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://huggingface.co/papers/2303.05499) by Shilong Liu, Zhaoyang Zeng, Tianhe Ren, Feng Li, Hao Zhang, Jie Yang, Qing Jiang, Chunyuan Li, Jianwei Yang, Hang Su, Jun Zhu, Lei Zhang.
 1. **[GroupViT](https://huggingface.co/docs/transformers/model_doc/groupvit)** (from UCSD, NVIDIA) released with the paper [GroupViT: Semantic Segmentation Emerges from Text Supervision](https://huggingface.co/papers/2202.11094) by Jiarui Xu, Shalini De Mello, Sifei Liu, Wonmin Byeon, Thomas Breuel, Jan Kautz, Xiaolong Wang.
 1. **[Helium](https://huggingface.co/docs/transformers/main/model_doc/helium)** (from the Kyutai Team) released with the blog post [Announcing Helium-1 Preview](https://kyutai.org/2025/01/13/helium.html) by the Kyutai Team.
diff --git a/package-lock.json b/package-lock.json
diff --git a/package.json b/package.json
@@ -1,6 +1,6 @@
 {
   "name": "@huggingface/transformers",
-  "version": "3.7.4",
+  "version": "3.7.5",
   "description": "State-of-the-art Machine Learning for the web. Run 🤗 Transformers directly in your browser, with no need for a server!",
   "main": "./src/transformers.js",
   "types": "./types/transformers.d.ts",
diff --git a/src/configs.js b/src/configs.js
@@ -119,6 +119,7 @@ function getNormalizedConfig(config) {
         case 'olmo2':
         case 'mobilellm':
         case 'granite':
+        case 'granitemoehybrid':
         case 'cohere':
         case 'mistral':
         case 'starcoder2':
diff --git a/src/env.js b/src/env.js
@@ -26,7 +26,7 @@ import fs from 'node:fs';
 import path from 'node:path';
 import url from 'node:url';
 
-const VERSION = '3.7.4';
+const VERSION = '3.7.5';
 
 // Check if various APIs are available (depends on environment)
 const IS_BROWSER_ENV = typeof window !== "undefined" && typeof window.document !== "undefined";
diff --git a/src/models.js b/src/models.js
@@ -1427,6 +1427,12 @@ export class PreTrainedModel extends Callable {
             processors.push(new ClassifierFreeGuidanceLogitsProcessor(generation_config.guidance_scale));
         }
 
+
+        if (generation_config.temperature === 0 && generation_config.do_sample) {
+          console.warn('`do_sample` changed to false because `temperature: 0` implies greedy sampling (always selecting the most likely token), which is incompatible with `do_sample: true`.');
+          generation_config.do_sample = false;
+        }
+
         if (generation_config.do_sample) {
             if (generation_config.temperature !== null && generation_config.temperature !== 1.0) {
                 processors.push(new TemperatureLogitsWarper(generation_config.temperature));
@@ -4653,6 +4659,12 @@ export class GraniteModel extends GranitePreTrainedModel { }
 export class GraniteForCausalLM extends GranitePreTrainedModel { }
 //////////////////////////////////////////////////
 
+//////////////////////////////////////////////////
+// GraniteMoeHybrid models
+export class GraniteMoeHybridPreTrainedModel extends PreTrainedModel { }
+export class GraniteMoeHybridModel extends GraniteMoeHybridPreTrainedModel { }
+export class GraniteMoeHybridForCausalLM extends GraniteMoeHybridPreTrainedModel { }
+//////////////////////////////////////////////////
 
 //////////////////////////////////////////////////
 // Cohere models
@@ -7841,6 +7853,7 @@ const MODEL_MAPPING_NAMES_DECODER_ONLY = new Map([
     ['olmo2', ['Olmo2Model', Olmo2Model]],
     ['mobilellm', ['MobileLLMModel', MobileLLMModel]],
     ['granite', ['GraniteModel', GraniteModel]],
+    ['granitemoehybrid', ['GraniteMoeHybridModel', GraniteMoeHybridModel]],
     ['cohere', ['CohereModel', CohereModel]],
     ['gemma', ['GemmaModel', GemmaModel]],
     ['gemma2', ['Gemma2Model', Gemma2Model]],
@@ -7951,6 +7964,7 @@ const MODEL_FOR_CAUSAL_LM_MAPPING_NAMES = new Map([
     ['olmo2', ['Olmo2ForCausalLM', Olmo2ForCausalLM]],
     ['mobilellm', ['MobileLLMForCausalLM', MobileLLMForCausalLM]],
     ['granite', ['GraniteForCausalLM', GraniteForCausalLM]],
+    ['granitemoehybrid', ['GraniteMoeHybridForCausalLM', GraniteMoeHybridForCausalLM]],
     ['cohere', ['CohereForCausalLM', CohereForCausalLM]],
     ['gemma', ['GemmaForCausalLM', GemmaForCausalLM]],
     ['gemma2', ['Gemma2ForCausalLM', Gemma2ForCausalLM]],

Original file line number	Diff line number	Diff line change
`@@ -1,6 +1,6 @@`
`1`	`1`
`2`	`2`
`3`		`-By default, Transformers.js uses [hosted pretrained models](https://huggingface.co/models?library=transformers.js) and [precompiled WASM binaries](https://cdn.jsdelivr.net/npm/@huggingface/transformers@3.7.4/dist/), which should work out-of-the-box. You can customize this as follows:`
	`3`	`+By default, Transformers.js uses [hosted pretrained models](https://huggingface.co/models?library=transformers.js) and [precompiled WASM binaries](https://cdn.jsdelivr.net/npm/@huggingface/transformers@3.7.5/dist/), which should work out-of-the-box. You can customize this as follows:`
`4`	`4`
`5`	`5`	`### Settings`
`6`	`6`
Original file line number	Diff line number	Diff line change
`@@ -1,6 +1,6 @@`
`1`	`1`	`{`
`2`	`2`	`"name": "@huggingface/transformers",`
`3`		`- "version": "3.7.4",`
	`3`	`+ "version": "3.7.5",`
`4`	`4`	`"description": "State-of-the-art Machine Learning for the web. Run 🤗 Transformers directly in your browser, with no need for a server!",`
`5`	`5`	`"main": "./src/transformers.js",`
`6`	`6`	`"types": "./types/transformers.d.ts",`