Background
On 16 November 2026, Hugging Face and OpenRouter announced a partnership that changes how requests to language models are routed through Hugging Face Inference Providers.[1] Inference Providers is the service through which Hugging Face sends model requests to hosting providers.
What changed
Hugging Face switched the routing of language-model requests for Inference Providers from its own custom router to OpenRouter's router.[1]
How provider selection works
OpenRouter picks a provider for each request based on live price, speed and reliability. If the first provider fails, it retries the request on another provider.[1]
Billing and keys
Language-model requests are now billed to the user's Hugging Face account or organization. Custom provider keys no longer work for those models.[1] Hugging Face says it adds no markup on provider prices.[1]
Supported APIs
The router supports OpenAI's Chat Completions and Responses APIs, as well as Anthropic's Messages API, for every chat model.[1]
What stays the same
Image, video and audio generation continue to go directly through Hugging Face Inference Providers.[1]
Why it matters
For developers, the change affects three things: which router handles language-model requests, which account is billed, and which keys work. Requests for media generation keep their existing path. Knowing which path a request takes is the practical point of the partnership.
Sources
- 1
Cite this page: Hugging Face and OpenRouter partner to route open-weight model requests. TellWorld, https://www.tellworld.net/news/hugging-face-openrouter-2026. Updated 2026-11-18