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

# CLI: Inspect Results

> Scores, failure explanations, fairness audits, and model summaries.

`R` is `--repo OWNER/NAME`, or another [scope](https://docs.selfbench.dev/cli/overview.md#conventions).

| Command | Does |
| - | - |
| `results R` | Pass rate and cost per configuration, and every task, hardest first |
| `task results RUN_ID TASK_ID R` | Every configuration on one task, with failure explanations and the fairness verdict |
| `task explain RUN_ID TASK_ID R [--config KEY] [--model M --credential ID] [--wait]` | Explain a task's unexplained failures |
| `task audit RUN_ID TASK_ID R --model M --credential ID [--wait]` | Judge whether the task is `fair`, `unfair`, or `unclear` |
| `summaries R` | The latest Model Summaries report |
| `summaries create R --model M --credential ID [--config KEY ...] [--wait]` | Summarize how configurations differ |

`task explain` uses a managed model where the deployment offers one. Audits and summaries need a saved model API key.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.