Metadata-Version: 2.4
Name: fusionkit-evals
Version: 0.10.1
Summary: Evaluation and Pareto analysis tools for fusionkit.
Keywords: fusionkit,model-fusion,llm,ensemble,evaluation,benchmarks
Author: Velum Labs
License-Expression: Apache-2.0
License-File: LICENSE
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Classifier: Operating System :: OS Independent
Requires-Dist: fusionkit-core==0.10.1
Requires-Dist: hyperkit==0.1.0
Requires-Dist: pydantic>=2.12.5
Requires-Dist: pyyaml>=6.0.3
Requires-Dist: typer>=0.27.0
Requires-Dist: pandas>=2.3.3 ; extra == 'evals'
Requires-Dist: datasets<6 ; extra == 'livecodebench'
Requires-Python: >=3.11
Project-URL: Homepage, https://fusionkit.velum-labs.com
Project-URL: Repository, https://github.com/velum-labs/handoffkit
Project-URL: Documentation, https://fusionkit.velum-labs.com
Provides-Extra: evals
Provides-Extra: livecodebench
Description-Content-Type: text/markdown

# fusionkit-evals

Evaluation and Pareto analysis helpers for FusionKit.

This package owns the canonical `fusionkit-bench` Typer app, benchmark runners,
tiny fixtures, public-benchmark tooling, prompt tuning, score analysis, and the
FusionKit HyperKit plugin. It depends on `fusionkit-core`, not the internal
`fusionkit` sidecar distribution.

Most users should start with `@fusionkit/cli`; install evaluation extras only when running benchmarks.

```sh
uv run --package fusionkit-evals fusionkit-bench --help
```

Docs: https://fusionkit.velum-labs.com
Repository: https://github.com/velum-labs/handoffkit
