A project, hydrated with additional information.
Sessions are also referred to as "Projects" in the UI.
The number of runs in the project.
The median (50th percentile) latency for the project.
The 99th percentile latency for the project.
The total number of tokens consumed in the project.
The total number of prompt tokens consumed in the project.
The total number of completion tokens consumed in the project.
The start time of the last run in the project.
Feedback stats for the project.
Summary feedback stats for the project.
Facets for the runs in the project.
The total estimated LLM cost associated with the completion tokens.
The estimated cost associated with the prompt (input) tokens.
The estimated cost associated with the completion tokens.
The median (50th percentile) time to process the first token.
The 99th percentile time to process the first token.
The error rate for the project.
End time of the run, if applicable.
Optional name for the run.
The reference dataset IDs this project's runs were generated on.
URL of this run within the app.
Optional metadata for the run.
Optional list of tags for the run.