Merge remote-tracking branch 'upstream/main'

Author: hanouticelina

docs/inference-providers/providers/cohere.md

@@ -56,6 +56,6 @@ Find out more about Chat Completion (VLM) [here](../tasks/chat-completion).
 
 <InferenceSnippet
     pipeline=image-text-to-text
-    providersMapping={ {"cohere":{"modelId":"CohereLabs/aya-vision-8b","providerModelId":"c4ai-aya-vision-8b"} } }
+    providersMapping={ {"cohere":{"modelId":"CohereLabs/aya-vision-32b","providerModelId":"c4ai-aya-vision-32b"} } }
 conversational />
 

docs/inference-providers/providers/fal-ai.md

@@ -64,6 +64,6 @@ Find out more about Text To Video [here](../tasks/text_to_video).
 
 <InferenceSnippet
     pipeline=text-to-video
-    providersMapping={ {"fal-ai":{"modelId":"Wan-AI/Wan2.1-T2V-14B","providerModelId":"fal-ai/wan-t2v"} } }
+    providersMapping={ {"fal-ai":{"modelId":"Lightricks/LTX-Video","providerModelId":"fal-ai/ltx-video"} } }
 />
 

docs/inference-providers/providers/hf-inference.md

@@ -39,13 +39,23 @@ If you are interested in deploying models to a dedicated and autoscaling infrast
 ## Supported tasks
 
 
+### Audio Classification
+
+Find out more about Audio Classification [here](../tasks/audio_classification).
+
+<InferenceSnippet
+    pipeline=audio-classification
+    providersMapping={ {"hf-inference":{"modelId":"firdhokk/speech-emotion-recognition-with-openai-whisper-large-v3","providerModelId":"firdhokk/speech-emotion-recognition-with-openai-whisper-large-v3"} } }
+/>
+
+
 ### Automatic Speech Recognition
 
 Find out more about Automatic Speech Recognition [here](../tasks/automatic_speech_recognition).
 
 <InferenceSnippet
     pipeline=automatic-speech-recognition
-    providersMapping={ {"hf-inference":{"modelId":"openai/whisper-large-v3-turbo","providerModelId":"openai/whisper-large-v3-turbo"} } }
+    providersMapping={ {"hf-inference":{"modelId":"openai/whisper-large-v3","providerModelId":"openai/whisper-large-v3"} } }
 />
 
 
@@ -65,7 +75,7 @@ Find out more about Chat Completion (VLM) [here](../tasks/chat-completion).
 
 <InferenceSnippet
     pipeline=image-text-to-text
-    providersMapping={ {"hf-inference":{"modelId":"meta-llama/Llama-3.2-11B-Vision-Instruct","providerModelId":"meta-llama/Llama-3.2-11B-Vision-Instruct"} } }
+    providersMapping={ {"hf-inference":{"modelId":"google/gemma-3-27b-it","providerModelId":"google/gemma-3-27b-it"} } }
 conversational />
 
 
@@ -75,7 +85,77 @@ Find out more about Feature Extraction [here](../tasks/feature_extraction).
 
 <InferenceSnippet
     pipeline=feature-extraction
-    providersMapping={ {"hf-inference":{"modelId":"kyutai/mimi","providerModelId":"kyutai/mimi"} } }
+    providersMapping={ {"hf-inference":{"modelId":"intfloat/multilingual-e5-large-instruct","providerModelId":"intfloat/multilingual-e5-large-instruct"} } }
+/>
+
+
+### Fill Mask
+
+Find out more about Fill Mask [here](../tasks/fill_mask).
+
+<InferenceSnippet
+    pipeline=fill-mask
+    providersMapping={ {"hf-inference":{"modelId":"google-bert/bert-base-uncased","providerModelId":"google-bert/bert-base-uncased"} } }
+/>
+
+
+### Image Classification
+
+Find out more about Image Classification [here](../tasks/image_classification).
+
+<InferenceSnippet
+    pipeline=image-classification
+    providersMapping={ {"hf-inference":{"modelId":"Falconsai/nsfw_image_detection","providerModelId":"Falconsai/nsfw_image_detection"} } }
+/>
+
+
+### Image Segmentation
+
+Find out more about Image Segmentation [here](../tasks/image_segmentation).
+
+<InferenceSnippet
+    pipeline=image-segmentation
+    providersMapping={ {"hf-inference":{"modelId":"mattmdjaga/segformer_b2_clothes","providerModelId":"mattmdjaga/segformer_b2_clothes"} } }
+/>
+
+
+### Object Detection
+
+Find out more about Object Detection [here](../tasks/object_detection).
+
+<InferenceSnippet
+    pipeline=object-detection
+    providersMapping={ {"hf-inference":{"modelId":"facebook/detr-resnet-50","providerModelId":"facebook/detr-resnet-50"} } }
+/>
+
+
+### Question Answering
+
+Find out more about Question Answering [here](../tasks/question_answering).
+
+<InferenceSnippet
+    pipeline=question-answering
+    providersMapping={ {"hf-inference":{"modelId":"deepset/roberta-base-squad2","providerModelId":"deepset/roberta-base-squad2"} } }
+/>
+
+
+### Summarization
+
+Find out more about Summarization [here](../tasks/summarization).
+
+<InferenceSnippet
+    pipeline=summarization
+    providersMapping={ {"hf-inference":{"modelId":"facebook/bart-large-cnn","providerModelId":"facebook/bart-large-cnn"} } }
+/>
+
+
+### Table Question Answering
+
+Find out more about Table Question Answering [here](../tasks/table_question_answering).
+
+<InferenceSnippet
+    pipeline=table-question-answering
+    providersMapping={ {"hf-inference":{"modelId":"google/tapas-base-finetuned-wtq","providerModelId":"google/tapas-base-finetuned-wtq"} } }
 />
 
 
@@ -108,3 +188,33 @@ Find out more about Text To Image [here](../tasks/text_to_image).
     providersMapping={ {"hf-inference":{"modelId":"black-forest-labs/FLUX.1-dev","providerModelId":"black-forest-labs/FLUX.1-dev"} } }
 />
 
+
+### Token Classification
+
+Find out more about Token Classification [here](../tasks/token_classification).
+
+<InferenceSnippet
+    pipeline=token-classification
+    providersMapping={ {"hf-inference":{"modelId":"dslim/bert-base-NER","providerModelId":"dslim/bert-base-NER"} } }
+/>
+
+
+### Translation
+
+Find out more about Translation [here](../tasks/translation).
+
+<InferenceSnippet
+    pipeline=translation
+    providersMapping={ {"hf-inference":{"modelId":"google-t5/t5-base","providerModelId":"google-t5/t5-base"} } }
+/>
+
+
+### Zero Shot Classification
+
+Find out more about Zero Shot Classification [here](../tasks/zero_shot_classification).
+
+<InferenceSnippet
+    pipeline=zero-shot-classification
+    providersMapping={ {"hf-inference":{"modelId":"facebook/bart-large-mnli","providerModelId":"facebook/bart-large-mnli"} } }
+/>
+

docs/inference-providers/providers/nebius.md

@@ -44,7 +44,7 @@ Find out more about Chat Completion (LLM) [here](../tasks/chat-completion).
 
 <InferenceSnippet
     pipeline=text-generation
-    providersMapping={ {"nebius":{"modelId":"deepseek-ai/DeepSeek-V3-0324","providerModelId":"deepseek-ai/DeepSeek-V3-0324-fast"} } }
+    providersMapping={ {"nebius":{"modelId":"Qwen/Qwen3-235B-A22B","providerModelId":"Qwen/Qwen3-235B-A22B"} } }
 conversational />
 
 

docs/inference-providers/register-as-a-provider.md

@@ -457,6 +457,28 @@ class MyNewProviderTaskProviderHelper(TaskProviderHelper):
 - Go to [tests/test_inference_providers.py](https://github.com/huggingface/huggingface_hub/blob/main/tests/test_inference_providers.py) and add static tests for overridden methods.
 
 
+## 6. Add provider documentation
+
+Create a dedicated documentation page for your provider within the Hugging Face documentation. This page should contain a concise description of your provider services, highlight the benefits for users, set expectations regarding performance or features, and include any relevant details such as pricing models or data retention policies. Essentially, provide any information that would be valuable to end users.
+
+Here's how to add your documentation page:
+
+-   Provide Your Logo: You can send your logo files (separate light and dark mode versions) directly to us. This is often the simplest way. Alternatively, if you prefer, you can open a PR in the [huggingface/documentation-images](https://huggingface.co/datasets/huggingface/documentation-images/tree/main/inference-providers/logos) repository. If you choose to open a PR:
+    *   Logos must be in `.png` format.
+    *   Name them `{provider-name}-light.png` and `{provider-name}-dark.png`.
+    *   Please ping `@Wauplin` and `@celinah` on the PR.
+-   Create the Documentation File:
+    *   Use an existing provider page as a template. For example, check out the template for [Fal AI](https://github.com/huggingface/hub-docs/blob/main/scripts/inference-providers/templates/providers/fal-ai.handlebars).
+    *   The file should be located under `scripts/inference-providers/templates/providers/{your-provider-name}.handlebars`.
+-   Submit the Documentation PR:
+    *   Add your new `{provider-name}.handlebars` file.
+    *   Update the [partners table](./index#partners) to include your company or product.
+    *   Update the `_toctree.yml` file in the `docs/inference-providers/` directory to include your new documentation page in the "Providers" section, maintaining alphabetical order.
+    *   Update the `scripts/inference-providers/scripts/generate.ts` file to include your provider in the `PROVIDERS_HUB_ORGS` and `PROVIDERS_URLS` constants, maintaining alphabetical order.
+    *   Run `pnpm install` (if you haven't already) and then `pnpm run generate` at the root of the `scripts/inference-providers` repository to generate the documentation. 
+    *    Commit all your changes, including the manually edited files (provider page, `_toctree.yml`, partners table) and the files generated by the script.
+    *   When you open the PR, please ping @Wauplin, @SBrandeis, @julien-c, and @hanouticelina for a review. If you need any assistance with these steps, please reach out – we're here to help you!
+
 ## FAQ
 
 **Question:** By default, in which order do we list providers in the settings page?

docs/inference-providers/tasks/audio-classification.md

@@ -29,13 +29,17 @@ For more details about the `audio-classification` task, check out its [dedicated
 
 ### Recommended models
 
+- [ehcalabres/wav2vec2-lg-xlsr-en-speech-emotion-recognition](https://huggingface.co/ehcalabres/wav2vec2-lg-xlsr-en-speech-emotion-recognition): An emotion recognition model.
 
 Explore all available models and find the one that suits you best [here](https://huggingface.co/models?inference=warm&pipeline_tag=audio-classification&sort=trending).
 
 ### Using the API
 
 
-There are currently no snippet examples for the **audio-classification** task, as no providers support it yet.
+<InferenceSnippet
+    pipeline=audio-classification
+    providersMapping={ {"hf-inference":{"modelId":"firdhokk/speech-emotion-recognition-with-openai-whisper-large-v3","providerModelId":"firdhokk/speech-emotion-recognition-with-openai-whisper-large-v3"}} }
+/>
 
 
 

docs/inference-providers/tasks/automatic-speech-recognition.md

@@ -38,7 +38,7 @@ Explore all available models and find the one that suits you best [here](https:/
 
 <InferenceSnippet
     pipeline=automatic-speech-recognition
-    providersMapping={ {"fal-ai":{"modelId":"openai/whisper-large-v3","providerModelId":"fal-ai/whisper"},"hf-inference":{"modelId":"openai/whisper-large-v3-turbo","providerModelId":"openai/whisper-large-v3-turbo"}} }
+    providersMapping={ {"fal-ai":{"modelId":"openai/whisper-large-v3","providerModelId":"fal-ai/whisper"},"hf-inference":{"modelId":"openai/whisper-large-v3","providerModelId":"openai/whisper-large-v3"}} }
 />
 
 

docs/inference-providers/tasks/chat-completion.md

@@ -24,6 +24,7 @@ This is a subtask of [`text-generation`](https://huggingface.co/docs/inference-p
 - [google/gemma-2-2b-it](https://huggingface.co/google/gemma-2-2b-it): A text-generation model trained to follow instructions.
 - [meta-llama/Meta-Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B-Instruct): Very powerful text generation model trained to follow instructions.
 - [microsoft/phi-4](https://huggingface.co/microsoft/phi-4): Powerful text generation model by Microsoft.
+- [Qwen/Qwen2.5-7B-Instruct-1M](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct-1M): Strong conversational model that supports very long instructions.
 - [Qwen/Qwen2.5-Coder-32B-Instruct](https://huggingface.co/Qwen/Qwen2.5-Coder-32B-Instruct): Text generation model used to write code.
 - [deepseek-ai/DeepSeek-R1](https://huggingface.co/deepseek-ai/DeepSeek-R1): Powerful reasoning based open large language model.
 
@@ -59,7 +60,7 @@ The API supports:
 
 <InferenceSnippet
     pipeline=text-generation
-    providersMapping={ {"cerebras":{"modelId":"meta-llama/Llama-3.3-70B-Instruct","providerModelId":"llama-3.3-70b"},"cohere":{"modelId":"CohereLabs/c4ai-command-a-03-2025","providerModelId":"command-a-03-2025"},"fireworks-ai":{"modelId":"Qwen/Qwen3-235B-A22B","providerModelId":"accounts/fireworks/models/qwen3-235b-a22b"},"hf-inference":{"modelId":"Qwen/Qwen3-235B-A22B","providerModelId":"Qwen/Qwen3-235B-A22B"},"hyperbolic":{"modelId":"deepseek-ai/DeepSeek-V3-0324","providerModelId":"deepseek-ai/DeepSeek-V3-0324"},"nebius":{"modelId":"deepseek-ai/DeepSeek-V3-0324","providerModelId":"deepseek-ai/DeepSeek-V3-0324-fast"},"novita":{"modelId":"Qwen/Qwen3-235B-A22B","providerModelId":"qwen/qwen3-235b-a22b-fp8"},"sambanova":{"modelId":"deepseek-ai/DeepSeek-V3-0324","providerModelId":"DeepSeek-V3-0324"},"together":{"modelId":"deepseek-ai/DeepSeek-R1","providerModelId":"deepseek-ai/DeepSeek-R1"}} }
+    providersMapping={ {"cerebras":{"modelId":"meta-llama/Llama-3.3-70B-Instruct","providerModelId":"llama-3.3-70b"},"cohere":{"modelId":"CohereLabs/c4ai-command-a-03-2025","providerModelId":"command-a-03-2025"},"fireworks-ai":{"modelId":"Qwen/Qwen3-235B-A22B","providerModelId":"accounts/fireworks/models/qwen3-235b-a22b"},"hf-inference":{"modelId":"Qwen/Qwen3-235B-A22B","providerModelId":"Qwen/Qwen3-235B-A22B"},"hyperbolic":{"modelId":"deepseek-ai/DeepSeek-V3-0324","providerModelId":"deepseek-ai/DeepSeek-V3-0324"},"nebius":{"modelId":"Qwen/Qwen3-235B-A22B","providerModelId":"Qwen/Qwen3-235B-A22B"},"novita":{"modelId":"Qwen/Qwen3-235B-A22B","providerModelId":"qwen/qwen3-235b-a22b-fp8"},"sambanova":{"modelId":"deepseek-ai/DeepSeek-V3-0324","providerModelId":"DeepSeek-V3-0324"},"together":{"modelId":"deepseek-ai/DeepSeek-R1","providerModelId":"deepseek-ai/DeepSeek-R1"}} }
 conversational />
 
 
@@ -69,7 +70,7 @@ conversational />
 
 <InferenceSnippet
     pipeline=image-text-to-text
-    providersMapping={ {"cerebras":{"modelId":"meta-llama/Llama-4-Scout-17B-16E-Instruct","providerModelId":"llama-4-scout-17b-16e-instruct"},"cohere":{"modelId":"CohereLabs/aya-vision-8b","providerModelId":"c4ai-aya-vision-8b"},"fireworks-ai":{"modelId":"meta-llama/Llama-4-Scout-17B-16E-Instruct","providerModelId":"accounts/fireworks/models/llama4-scout-instruct-basic"},"hf-inference":{"modelId":"meta-llama/Llama-3.2-11B-Vision-Instruct","providerModelId":"meta-llama/Llama-3.2-11B-Vision-Instruct"},"hyperbolic":{"modelId":"Qwen/Qwen2.5-VL-7B-Instruct","providerModelId":"Qwen/Qwen2.5-VL-7B-Instruct"},"nebius":{"modelId":"google/gemma-3-27b-it","providerModelId":"google/gemma-3-27b-it-fast"},"novita":{"modelId":"meta-llama/Llama-4-Scout-17B-16E-Instruct","providerModelId":"meta-llama/llama-4-scout-17b-16e-instruct"},"sambanova":{"modelId":"meta-llama/Llama-4-Scout-17B-16E-Instruct","providerModelId":"Llama-4-Scout-17B-16E-Instruct"},"together":{"modelId":"meta-llama/Llama-4-Scout-17B-16E-Instruct","providerModelId":"meta-llama/Llama-4-Scout-17B-16E-Instruct"}} }
+    providersMapping={ {"cerebras":{"modelId":"meta-llama/Llama-4-Scout-17B-16E-Instruct","providerModelId":"llama-4-scout-17b-16e-instruct"},"cohere":{"modelId":"CohereLabs/aya-vision-32b","providerModelId":"c4ai-aya-vision-32b"},"fireworks-ai":{"modelId":"meta-llama/Llama-4-Scout-17B-16E-Instruct","providerModelId":"accounts/fireworks/models/llama4-scout-instruct-basic"},"hf-inference":{"modelId":"google/gemma-3-27b-it","providerModelId":"google/gemma-3-27b-it"},"hyperbolic":{"modelId":"Qwen/Qwen2.5-VL-7B-Instruct","providerModelId":"Qwen/Qwen2.5-VL-7B-Instruct"},"nebius":{"modelId":"google/gemma-3-27b-it","providerModelId":"google/gemma-3-27b-it-fast"},"novita":{"modelId":"meta-llama/Llama-4-Scout-17B-16E-Instruct","providerModelId":"meta-llama/llama-4-scout-17b-16e-instruct"},"sambanova":{"modelId":"meta-llama/Llama-4-Scout-17B-16E-Instruct","providerModelId":"Llama-4-Scout-17B-16E-Instruct"},"together":{"modelId":"meta-llama/Llama-4-Scout-17B-16E-Instruct","providerModelId":"meta-llama/Llama-4-Scout-17B-16E-Instruct"}} }
 conversational />
 
 
@@ -120,6 +121,11 @@ conversational />
 | **&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;(#2)** | _object_ |  |
 | **&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;type*** | _enum_ | Possible values: regex. |
 | **&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;value*** | _string_ |  |
+| **&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;(#3)** | _object_ |  |
+| **&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;type*** | _enum_ | Possible values: json_schema. |
+| **&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;value*** | _object_ |  |
+| **&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;name** | _string_ | Optional name identifier for the schema |
+| **&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;schema*** | _unknown_ | The actual JSON schema definition |
 | **seed** | _integer_ |  |
 | **stop** | _string[]_ | Up to 4 sequences where the API will stop generating further tokens. |
 | **stream** | _boolean_ |  |

docs/inference-providers/tasks/feature-extraction.md

@@ -29,6 +29,7 @@ For more details about the `feature-extraction` task, check out its [dedicated p
 
 ### Recommended models
 
+- [thenlper/gte-large](https://huggingface.co/thenlper/gte-large): A powerful feature extraction model for natural language processing tasks.
 
 Explore all available models and find the one that suits you best [here](https://huggingface.co/models?inference=warm&pipeline_tag=feature-extraction&sort=trending).
 
@@ -37,7 +38,7 @@ Explore all available models and find the one that suits you best [here](https:/
 
 <InferenceSnippet
     pipeline=feature-extraction
-    providersMapping={ {"hf-inference":{"modelId":"kyutai/mimi","providerModelId":"kyutai/mimi"},"sambanova":{"modelId":"intfloat/e5-mistral-7b-instruct","providerModelId":"E5-Mistral-7B-Instruct"}} }
+    providersMapping={ {"hf-inference":{"modelId":"intfloat/multilingual-e5-large-instruct","providerModelId":"intfloat/multilingual-e5-large-instruct"},"sambanova":{"modelId":"intfloat/e5-mistral-7b-instruct","providerModelId":"E5-Mistral-7B-Instruct"}} }
 />
 
 

docs/inference-providers/tasks/fill-mask.md

@@ -24,13 +24,17 @@ For more details about the `fill-mask` task, check out its [dedicated page](http
 
 ### Recommended models
 
+- [FacebookAI/xlm-roberta-base](https://huggingface.co/FacebookAI/xlm-roberta-base): A multilingual model trained on 100 languages.
 
 Explore all available models and find the one that suits you best [here](https://huggingface.co/models?inference=warm&pipeline_tag=fill-mask&sort=trending).
 
 ### Using the API
 
 
-There are currently no snippet examples for the **fill-mask** task, as no providers support it yet.
+<InferenceSnippet
+    pipeline=fill-mask
+    providersMapping={ {"hf-inference":{"modelId":"google-bert/bert-base-uncased","providerModelId":"google-bert/bert-base-uncased"}} }
+/>