Hugging Face ditches your API keys to become an AI traffic cop
Original headline: Hugging Face now routes model requests through OpenRouter for the best price or speed — the-decoder.com
More in AI and machine learning news
Hugging Face has quietly changed a core part of its developer platform. Its Inference Providers service now automatically routes all large language model (LLM) requests through a middleman called OpenRouter. This router picks which AI model provider—like Anthropic, Google, or a smaller open-source host—gets your request, deciding based on a mix of price, speed, and reliability. Developers can add suffixes like ':fastest' or ':cheapest' to prioritize one factor, but the default is now this automated, hands-off routing.
The practical impact is a shift in control away from developers. You can no longer use your own API keys directly with language model providers on Hugging Face; that option is now limited to image and video models. While Hugging Face promises no markup and cites OpenRouter's stats showing fewer errors and faster responses, the move locks developers into a new layer of abstraction. The takeaway is familiar: another platform consolidates its gatekeeper role, simplifying things for beginners while quietly removing options for power users who preferred to manage their own connections.
Read the original at the-decoder.com
Satirical commentary on real tech news, summarised with AI. Read the original source for the facts. How ZeroDaily works
