Scaleway, a European cloud provider, has joined Hugging Face's Inference Providers ecosystem, allowing developers to run serverless inference on models hosted on Hugging Face directly from model pages and client libraries. This integration expands options for serverless inference with both open-weight and commercial models, making it easier for developers to access powerful AI models without handling servers.

What Happened

On September 19, 2025, Hugging Face announced that Scaleway is now an Inference Provider supported on the Hub. This partnership enables developers to directly select Scaleway as a provider for running serverless inference on models hosted on Hugging Face. They can choose between using Hugging Face's routing or their own Scaleway API keys, giving them more flexibility to test, benchmark, and deploy inference at scale.

By becoming an Inference Provider on Hugging Face, Scaleway joins a global ecosystem where openness, flexibility, and performance are key. This integration expands the range of providers available to the Hugging Face community, while giving developers the opportunity to run inference on a European cloud fully aligned with local standards and regulations.

Background and Context

Hugging Face's Inference API is a service that handles everything related to deploying and running thousands of pretrained deep learning models from the Hugging Face Hub. These include advanced large language models (LLMs) like LLaMA and BERT, as well as vision models like CLIP, or models for specific tasks like translation, summarization, question answering, and text classification.

However, running these powerful AI models requires significant infrastructure and resources. To simplify this process, Hugging Face introduced the concept of Inference Providers – cloud platforms that run the background work for model setups. These providers include big cloud companies like AWS and Google Cloud Platform (GCP), as well as specialized providers like CoreWeave.

Now, with Scaleway joining the list of Inference Providers, developers have a faster and cheaper way to run models without handling servers. This is particularly important for creators and developers who want to make AI setups simpler using serverless infrastructure.

Why It Matters

This integration matters significantly for several reasons:

  • Faster Inference Times: Scaleway's cold starts average under 30ms, with inference results delivered in under one second. This is a significant improvement over other cloud providers.
  • Cheaper Pricing: Serverless inference on Scaleway costs ~$0.0020 per second – significantly cheaper than AWS or GCP.
  • Data Residency Concerns: EU-based Scaleway supports GDPR and data residency concerns better than many U.S.-centric providers.
  • Easy Deployment: Over 80 models were deployed on Scaleway within 30 days of joining Hugging Face Inference Providers, demonstrating the ease of use and deployment.
  • No-Code Platforms: Serverless inference takes care of all infrastructure, letting no-code platforms add LLMs easily.

What Comes Next

The integration of Scaleway as an Inference Provider on Hugging Face is a significant step forward for the industry. As developers continue to explore and deploy AI models, this partnership will provide them with a faster, cheaper, and more secure way to run inference at scale.

Key Facts

  • Scaleway is now an Inference Provider on Hugging Face's Hub.
  • Developers can directly select Scaleway as a provider for running serverless inference on models hosted on Hugging Face.
  • Scaleway supports GDPR and data residency concerns better than many U.S.-centric providers.
  • Serverless inference on Scaleway costs ~$0.0020 per second – significantly cheaper than AWS or GCP.
  • Over 80 models were deployed on Scaleway within 30 days of joining Hugging Face Inference Providers.