Skip to content

armgit / janas

Public Imported from GitHub
0

37 tok/s from Qwen3-Next-80B (48 GB) on a 32 GB laptop, 24 without drafts vs 2.8 for llama.cpp; reading 62 vs 16 tok/s. Faster than llama.cpp on all 13 models compared in Q4_K_M. LLMs bigger than your RAM, pure C, CPU or GPU.

1.5 KB
Raw Blame
<!doctype html>
<!-- SPDX-License-Identifier: GPL-3.0-or-later
     Copyright (C) 2026 Maurizio Cammalleri -->
<html lang="en">
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>Janas-LLM in Node.js</title>
<link rel="stylesheet" href="/style.css">
<script type="module" src="/app/web/app.js"></script>
</head>
<body>
<header>
  <div class="brand">
    <span class="logo">J</span>
    <div>
      <div id="model-name" class="name">…</div>
      <div id="model-desc" class="desc">the model is opening</div>
    </div>
  </div>
  <button id="new" type="button" class="ghost">New chat</button>
</header>

<main id="log" aria-live="polite"></main>

<footer>
  <details id="settings">
    <summary>Settings</summary>
    <label for="system">System message</label>
    <textarea id="system" rows="3"></textarea>
    <div class="row">
      <label id="thinking-row" hidden><input id="thinking" type="checkbox"> reasoning before the answer</label>
      <label>temperature
        <input id="temperature" type="range" min="0" max="1.5" step="0.1" value="0.7">
        <output id="temperature-value"></output>
      </label>
    </div>
  </details>
  <div class="composer">
    <textarea id="input" rows="2" placeholder="Write a message (Enter sends, Shift+Enter a new line)"></textarea>
    <button id="send" type="button">Send</button>
    <button id="stop" type="button" class="ghost" hidden>Stop</button>
  </div>
  <div id="versions" class="versions"></div>
</footer>
</body>
</html>