README / index.html
K00lKidd's picture
Organisation card
4e70b7a verified
Raw History Blame Contribute Delete
1.79 kB
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>Debussy Labs</title>
</head>
<body>
<!-- Hugging Face shows this file as the organisation card on huggingface.co/debussylabs. It strips CSS and paragraph
margins, so the spacing between sections is done with <br>. Keep it simple HTML. -->
<p align="center"><img src="https://huggingface.co/spaces/debussylabs/README/resolve/main/debussy-labs-banner.png" width="327" height="86" alt="Debussy Labs"></p>
<br>
<p><b>Debussy Labs</b> builds fast, honest decision models for AI agents.</p>
<br>
<p>Our first model, <b>Debussy</b>, is an open <i>System 1</i> decision model. It makes the quick calls an agent needs
(is this request safe, should I call this tool, do I actually know this) in a single pass on an ordinary CPU, and it
says “I don't know” when it doesn't.</p>
<br>
<p><b>Models</b></p>
<ul>
<li><a href="https://huggingface.co/debussylabs/debussy"><b>debussylabs/debussy</b></a>: safety judgements, tool-call routing
and knowledge questions with “I don't know”. Apache-2.0.</li>
</ul>
<br>
<p><b>Estimated Decision Index ≈ 63.7</b> on the Artificial Analysis Decision Index (our own estimate, not an official
score). Built and evaluated on a 4-core CPU, with no GPU and no API key.</p>
<br>
<p>🌐 <a href="https://debussylabs.com">debussylabs.com</a> · 💻 <a href="https://github.com/abutauskas/Debussy">GitHub</a> ·
📰 <a href="https://debussylabs.com/news/introducing-debussy/">Introducing Debussy</a> ·
📦 <a href="https://debussylabs.com/releases/">Releases</a></p>
<br>
<p><small>Built with help from Claude, Anthropic's AI assistant. Debussy is not an Anthropic model.</small></p>
</body>
</html>