DeepInfra is now a supported Inference Provider on the Hugging Face Hub, offering serverless AI inference with support for conversational and text-generation tasks.

DeepInfra joins Hugging Face's ecosystem of Inference Providers, enabling developers to access models like DeepSeek V4, Kimi-K2.6, and GLM-5.1 directly from model pages and through client SDKs (Python and JavaScript).

Users can authenticate via Hugging Face token for routed requests (billed at standard provider rates with no markup) or use their own DeepInfra API key for direct requests (billed by DeepInfra). Hugging Face PRO users receive $2 worth of Inference credits monthly across providers.

Initial launch supports conversational and text-generation tasks on open-weight LLMs; support for text-to-image, text-to-video, embeddings, and other tasks will roll out soon.