Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.evarisai.com/llms.txt

Use this file to discover all available pages before exploring further.

Datasets are the source inputs for evaluation runs.

Typical dataset flow

  1. Create/import dataset samples.
  2. Attach datasets to an evaluation suite.
  3. Execute runs and inspect sample-level outputs.

Good dataset practices

  • Keep schema consistent for stable comparisons.
  • Version datasets when changing prompts/tasks.
  • Start small for fast iteration, then scale.