Sign inGet started
Repository · GitHub

tatsu-lab/alpaca_eval

An automatic evaluator for instruction-following language models. Human-validated, high-quality, cheap, and fast.

Ownertatsu-lab
License (as declared)Apache-2.0
Stars2,020
Forks314
Main languageJupyter Notebook
Created25 May 2023
Last updated at source9 August 2025
jupyter notebookdeep-learningevaluationfoundation-modelsinstruction-followinglarge-language-modelsleaderboardnlprlhf
Embed this entry
<iframe src="https://silytics.si/embed/item/df69f587-a68e-4dae-b16e-3322af4512d3" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Free, no tracking. More about feeds and embeds.

More from tatsu-lab

Silytics shows public metadata from GitHub and links to the original. Information may be incomplete or out of date. Last refreshed 15 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.