> ## Documentation Index
> Fetch the complete documentation index at: https://docs.praxa.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Benchmark data and reproduction

> Trace the published charts to versioned evidence, regenerate them locally, and preserve the pilot's claim boundary.

The charts in these docs are generated from a checked, machine-readable summary
of the public <code>preprint-v1.1.0</code> artifacts.

## Trace the evidence

```mermaid theme={null}
flowchart LR
  Release["preprint-v1.1.0 release"] --> Raw["Versioned JSON artifacts"]
  Raw --> Summary["Checked public summary"]
  Summary --> SVG["Accessible SVG charts"]
  Summary --> Tables["Adjacent data tables"]
```

| Artifact                                                                                               | Purpose                                            |
| ------------------------------------------------------------------------------------------------------ | -------------------------------------------------- |
| [Preprint release](https://github.com/praxa-labs/praxa-benchmarks/releases/tag/preprint-v1.1.0)        | Versioned publication boundary                     |
| [Paper data directory](https://github.com/praxa-labs/praxa-benchmarks/tree/preprint-v1.1.0/paper/data) | Public evidence inputs                             |
| [Local summary](/benchmarks/data/terminal-bench-summary.json)                                          | Values used by this docs build                     |
| [Outcomes chart](/images/benchmarks/terminal-bench-outcomes.svg)                                       | Trial-result visualization                         |
| [Resource chart](/images/benchmarks/terminal-bench-resources.svg)                                      | Token, step, verification, and rollback comparison |
| [Coverage chart](/images/benchmarks/repository-coverage.svg)                                           | Repository-local coverage visualization            |

## Regenerate the docs artifacts

From the docs repository:

```bash theme={null}
npm run docs:generate
npm run docs:generate:check
npm run docs:verify
```

The check mode fails if a generated product guide, service-status page, or
benchmark chart differs from its checked data source.

## Reproduce the pilot

Follow the protocol and environment instructions in the public benchmark
repository. Record:

* exact source and package revisions;
* model and provider configuration;
* task and attempt inventory;
* budgets and stop conditions;
* environment image and cache state;
* every exclusion and missing trial;
* raw outcomes before aggregate calculation.

Do not compare a new run to the published pilot unless the comparison
conditions are compatible. A reproduction with a different model or environment
is still useful, but it is a new experiment.

<Warning>
  Never put provider credentials, customer data, private traces, or secret
  prompts in a public reproduction artifact.
</Warning>