Skip to main content

Documentation Index

Fetch the complete documentation index at: https://narev.ai/docs/llms.txt

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

​
How does benchmarking with Narev work?

There are three steps to getting your benchmark up and running.
1

Create a dataset and define a quality metric

This dataset contains all prompts used in the benchmark. Here, you also define how to evaluate responses.
2

Add variants to your benchmark

Variants are a combination of:
  • model and provider (for example, DeepSeek R1 from OpenRouter)
  • system prompt
  • parameters
You can add as many variants as needed to determine the most cost efficient model.
3

Run the benchmark and interpret the results

4

Optional: publish benchmark to Hub

Narev Hub is a community of public benchmarks created by users.