Hugging Face adds support for Featherless AI inference

Featherless AI has joined Hugging Face Hub as a new Inference Provider.

Featherless AI offers a vast model catalog—including the latest open source models from DeepSeek, Meta, Google, Qwen, and more—via a Serverless approach. Users can instantly leverage a wide range of models without the burden of infrastructure management.

Key Features and Usage:

• SDK Integration: Easily integrate into existing code via the Python (huggingface_hub) and JavaScript (@huggingface/inference) SDKs.
• Flexible Billing: Supports both the Custom Key method, where users pay the provider directly using their own API key, and the Routed by HF method, where billing goes through a Hugging Face account.
• UI Integration: Users can check providers within widgets on Hugging Face model pages and prioritize their preferred providers according to their settings.