capabilities.work

Scrape TGI Prometheus metrics

One step with Hugging Face. We can help you get access.

Look something up

Scrape TGI Prometheus metrics

with Hugging Face

Use Hugging Face: its Hugging Face Text Generation Inference Info API lets you scrape TGI Prometheus metrics in a single step (GET /metrics). capabilities.work can help you get signed up and connected.

  • Can I scrape queue size and token throughput from a TGI server with Prometheus?
  • Where are the request count and latency metrics for Text Generation Inference exposed?
Say it to your AI assistant
  • Fetch the Prometheus metrics from the TGI server.
  • Show me TGI queue size and token throughput metrics.
For your developer

GET /metrics in Hugging Face Text Generation Inference Info API. Full spec on APIs.io.

More you can do with Hugging Face

Preview the first rows of a dataset splitPage through rows of a dataset splitList Parquet files via the Dataset ViewerGet a dataset's row count and byte sizeGet column statistics for a dataset splitGet a dataset's Croissant metadataCheck if a dataset works in the Dataset ViewerList a dataset's subsets and splitsSearch datasets on the HubGet details about one datasetList a dataset's auto-converted Parquet filesList dataset tags grouped by type