ARTFEED — Contemporary Art Intelligence

DeepInfra Joins Hugging Face Inference Providers

ai-technology · 2026-04-29

DeepInfra, a platform for serverless AI inference, has become a recognized Inference Provider on the Hugging Face Hub. It features affordable token pricing and boasts a library of more than 100 models, which includes LLMs, text-to-image, text-to-video, and embeddings. Currently, DeepInfra caters to conversational and text-generation applications on Hugging Face, providing access to models such as DeepSeek V4, Kimi-K2.6, and GLM-5.1. Users can utilize custom API keys or direct their requests through Hugging Face, with PRO users benefiting from $2 in monthly inference credits. This integration can be accessed through Hugging Face SDKs for both Python and JavaScript, along with several agent harnesses.

Key facts

  • DeepInfra is now a supported Inference Provider on the Hugging Face Hub.
  • DeepInfra offers serverless AI inference with cost-effective pricing per token.
  • Catalog includes over 100 models: LLMs, text-to-image, text-to-video, embeddings.
  • Initial support for conversational and text-generation tasks on Hugging Face.
  • Models available include DeepSeek V4, Kimi-K2.6, GLM-5.1.
  • Users can set custom API keys or route requests through Hugging Face.
  • PRO users get $2 worth of Inference credits monthly.
  • Integration works with Hugging Face SDKs (Python and JavaScript) and agent harnesses.

Entities

Institutions

  • DeepInfra
  • Hugging Face
  • Hugging Face Hub
  • Hugging Face SDKs
  • Python
  • JavaScript
  • Pi
  • OpenCode
  • Hermes Agents
  • OpenClaw

Sources