Hugging Face is launching a paid service that automates the creation of chatbots and AI systems, specifically designed to streamline development with open-source models.
Hugging Face offers a variety of open-source models on its platform. You can find models from major players such as Google (Gemma) and Meta (Llama), among many others. Developers can then use those shared models to build their own AI tools.
According to Hugging Face, developers often need additional support in this process. The company notes that building a chatbot or AI tool and choosing the right underlying chip infrastructure is a time-consuming task.
Offerings via AWS and Google Cloud
The help comes from HUGS, which is short for Hugging Face for Generative AI Services. The tool was developed by Hugging Face in partnership with Amazon and Google. That is why the offering is distributed through these companies’ cloud offerings for one dollar per hour per container.
Users also have the option to download the tool for on-premises deployment, addressing corporate data privacy concerns and making it suitable for heavily regulated industries such as finance and healthcare.
Regarding hardware compatibility, HUGS can currently work with Nvidia and AMD GPUs with certainty. Support for AWS Inferentia and Google TPUs will soon follow. The range of open-source models was also limited to thirteen popular LLMs at launch. These include models from Meta, Nous Research, Mistral AI, Google, and Qwen.
Also read: Google helps developers choose the best Gemini or Gemma model