- Open models on Hugging Face are callable via OpenRouter's unified API without per-provider integration
- Requests auto-route by price, speed and uptime with automatic failover
- Traces can be broadcast into Hugging Face Storage for evals and fine-tuning
AI CURATED HOT STORY · 11/16/2026
Hugging Face and OpenRouter Unify Open-Source Model Access
Hugging Face and OpenRouter have partnered to allow users to directly call open-weight models hosted by providers on the Hub using their Hugging Face token. OpenRouter routes requests to over 80 providers based on real-time measurements, lowering the barrier to deploying and calling open-source models.
Users simply link their OpenRouter account with Hugging Face to use the same token and API, eliminating the need to integrate with each inference provider separately. OpenRouter automatically selects the optimal provider based on price, speed, and availability, with automatic failover. Additionally, request traces can be written to Hugging Face Storage via Broadcast for evaluation and fine-tuning.
This collaboration bridges the gap between hosting and calling open-source models, significantly reducing the barrier for developers and accelerating the adoption of open-source models in various applications, while fostering competition and consolidation in the inference services market.
Hourly heat curve
29 fully observed hoursThe line compares only sources observed throughout; gap hours are interpolated to keep the trend continuous.
Coverage timeline
Total 2 reports · Latest first- OpenRouter (@OpenRouter)T1·58 pts
- Hugging Face BlogT1·62 pts
- Hugging Face tokens can directly call provider-served open-weight models on the Hub
- OpenRouter routes requests across 80+ providers based on live measurements
- Users must connect their OpenRouter account to Hugging Face to enable it