Crandore Hub

vitals

Large Language Model Evaluation

A port of 'Inspect', a widely adopted 'Python' framework for large language model evaluation. Specifically aimed at 'ellmer' users who want to measure the effectiveness of their large language model-based products, the package supports prompt engineering, tool usage, multi-turn dialog, and model graded evaluations.

README

## Package system files

**test/**

The package uses a number of cached objects during testing in `test/`.

The `.json` files in `inspect-example/logs` are the resulting log file from running evals in `inst/test/inspect`, e.g. with: 

```bash
inspect eval inst/test/inspect/basics.py  --model anthropic/claude-sonnet-4-5 --log-format=json
```

...or:

```bash
inspect eval test/inspect/tools.py  --model anthropic/claude-sonnet-4-5 --log-format=json
```

**dist/**

`/dist` is a bundled version of the Inspect viewer. (See [here](https://github.com/UKGovernmentBEIS/inspect_ai/blob/88d1cd98041a245c1d0cca4536d60e3244630b78/src/inspect_ai/_view/www/README.md) for more information.)

**regenerate-example-objects.R**

The package defines a function `regenerate_example_objects()` in the source that sources the script `inst/regenerate-example-objects.R`.

Versions across snapshots

VersionRepositoryFileSize
0.2.0 rolling linux/jammy R-4.5 vitals_0.2.0.tar.gz 2.8 MiB
0.2.0 rolling linux/noble R-4.5 vitals_0.2.0.tar.gz 2.8 MiB
0.2.0 rolling source/ R- vitals_0.2.0.tar.gz 2.7 MiB
0.2.0 latest linux/jammy R-4.5 vitals_0.2.0.tar.gz 2.8 MiB
0.2.0 latest linux/noble R-4.5 vitals_0.2.0.tar.gz 2.8 MiB
0.2.0 latest source/ R- vitals_0.2.0.tar.gz 2.7 MiB
0.2.0 2026-04-26 source/ R- vitals_0.2.0.tar.gz 2.7 MiB
0.2.0 2026-04-23 source/ R- vitals_0.2.0.tar.gz 2.7 MiB
0.2.0 2026-04-09 windows/windows R-4.5 vitals_0.2.0.zip 2.8 MiB

Dependencies (latest)

Imports

Suggests