AI BriefingHugging Face Integrates Baseten
Key point
Hugging Face has added Baseten as an Inference Provider to support serverless model inference.
Details
Hugging Face has added Baseten as an Inference Provider, enabling Baseten-based serverless inference on Hub model pages and via the Python and JavaScript SDKs.
Initially, conversational and text generation tasks are supported, allowing open-weight LLMs such as Kimi K3, DeepSeek V4 Flash, and GLM-5.2 to be invoked through Baseten. Additional task types are expected to be supported in the future.
Users can make requests in two ways:
- Register their own Baseten API key to call the provider directly
- Have Hugging Face route the requests and bill the costs to their Hugging Face account
The Hugging Face SDK supports huggingface_hub version 1.26.1 or higher and @huggingface/inference for JavaScript, allowing calls via an OpenAI-compatible API using the model-name:baseten format. Integration with major Agent Harnesses such as Pi, OpenCode, Hermes Agents, and OpenClaw is also provided.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.