Hugging Face
Hugging Face's Inference Providers router, fanned out across competing serving providers behind one chat-completions endpoint.
Where Hugging Face lands
Composite scores out of 10, from benchmark runs where every provider faces the identical task corpus. Rank is within that capability's field. Full method: how we measure.
Hugging Face sits outside the web-access benchmark — the search/scrape/crawl corpora don't exercise a model inference API, so this page carries catalog facts without a scorecard.
The honest pitch
Pick it to reach open models served by whichever backing provider is cataloged for them, without separate accounts or price lists per provider.
The leanest of the model-inference routes: no files, batches or fine-tuning surface, and no access to the router's own fastest/cheapest/preferred routing — chat completions only, pinned to specific cataloged (model, provider) pairs.
About Hugging Face
Hugging Face’s router serves each model through several competing backing providers whose prices differ, so the route pins every admitted model to a specific (model, provider) pair with its own verified rate rather than letting the router pick one invisibly. A bare model id resolves to its pinned default provider; an explicit pairing is honored only if that exact combination is cataloged, and the router’s own fastest/cheapest/preferred routing policies are declined because they would make the price unpredictable. The admitted surface is stateless chat completions only — no stored files, batches or fine-tuning — and an account can register its own Hugging Face token to bill that provider directly.
Reaching it through NativePort
Identical paths, parameters and responses to Hugging Face's own documentation — the gateway holds the upstream credential and meters your balance at Hugging Face's real published usage price.
curl https://api.nativeport.ai/huggingface/<native-path> \ -H "Authorization: Bearer $NATIVEPORT_API_KEY"