index.html

<!doctype html>
<html>
	<head>
		<meta charset="utf-8" />
		<meta name="viewport" content="width=device-width" />
		<title>Distributed LLM Inference Platform</title>
		<link rel="stylesheet" href="style.css" />
	</head>
	<body>
		<div class="card">
			<h1>Distributed LLM Inference Platform</h1>
			<p>Multi-GPU LLM serving: one model worker per GPU, with a load balancer that routes each request to the least-busy worker.</p>
			<p>Work in progress. Throughput and latency benchmarks will be published here.</p>
			<p>Code: <code>server/worker.py</code> · <code>server/load_balancer.py</code></p>
		</div>
	</body>
</html>