llm-test-bench/settings.html at main · Techdread/llm-test-bench · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
<!doctype html>
<html lang="en" data-theme="dark" data-distribution="public">
<head>
  <meta charset="utf-8">
  <meta name="viewport" content="width=device-width, initial-scale=1">
  <meta name="description" content="Connect a data folder and local AI models to the projects on this site.">
  <title>Settings · Peter Edwards</title>
  <link rel="canonical" href="https://neuroviz.uk/settings">
  <link rel="icon" href="favicon.svg" type="image/svg+xml">
  <link rel="apple-touch-icon" href="apple-touch-icon.png">
  <meta name="theme-color" content="#080b12">
  <link rel="stylesheet" href="site.css?v=3b2749e850">
  <script src="plexus.js?v=af28b17aab" defer></script>
  <script type="module" src="settings.js?v=40ad903ef1"></script>
</head>
<body>
  <header class="site-header">
    <a class="brand" href="./" aria-label="Peter Edwards — home"><span class="brand-mark">PE</span><span>Peter Edwards</span></a>
    <nav aria-label="Main navigation">
      <a href="./#projects">Projects</a>
      <a href="showcase.html">Showcase</a>
      <a href="./#about">About</a>
      <a href="privacy.html">Privacy</a>
      <a href="https://github.com/Techdread/llm-test-bench" data-repository-link>GitHub</a>
    </nav>
  </header>

  <main class="settings-page">
    <p class="eyebrow">Local workspace setup</p>
    <h1>Connect your tools</h1>
    <p class="settings-intro">Choose a data folder and connect an optional local model server. Your files, endpoint details, and generated work stay on your device.</p>

    <section class="settings-card" aria-labelledby="folder-title">
      <div class="setting-heading">
        <div>
          <p class="setting-label">Data root</p>
          <h2 id="folder-title">Your working folder</h2>
        </div>
        <span class="local-badge">Local only</span>
      </div>

      <div class="folder-status" id="folder-status" role="status" aria-live="polite">
        <span class="status-dot" aria-hidden="true"></span>
        <div>
          <strong id="status-title">Checking folder access…</strong>
          <p id="status-detail">Looking for a folder previously connected to this website.</p>
        </div>
      </div>

      <button class="button primary folder-button" id="choose-folder" type="button" disabled>Choose folder</button>
      <p class="setting-help" id="browser-help">Chrome or Edge will show a secure system folder picker. You can revoke access later in your browser's site settings.</p>
    </section>

    <section class="settings-card model-settings-card" aria-labelledby="models-title">
      <div class="setting-heading">
        <div>
          <p class="setting-label">Local AI</p>
          <h2 id="models-title">Your local model server</h2>
        </div>
        <span class="local-badge">Direct connection</span>
      </div>

      <p class="setting-help model-intro">The browser talks directly to your own LM Studio or OpenAI-compatible endpoint. Requests are not relayed through this site.</p>

      <ol class="model-setup-steps">
        <li>Start the server and load a model. For LM Studio, enable CORS in Developer settings or run <code>lms server start --cors</code>.</li>
        <li>Add the endpoint below—LM Studio normally uses <code>http://localhost:1234</code>.</li>
        <li>Click <strong>Add &amp; test</strong>, then choose <strong>Allow</strong> if Chrome or Edge asks for local-network access.</li>
      </ol>

      <form class="model-endpoint-form" id="model-endpoint-form">
        <label>
          <span>Server type</span>
          <select id="model-provider-type">
            <option value="lmstudio">LM Studio / OpenAI-compatible</option>
            <option value="unsloth-studio">Unsloth Studio</option>
          </select>
        </label>
        <label>
          <span>Display name</span>
          <input id="model-provider-name" value="LM Studio" autocomplete="off">
        </label>
        <label class="endpoint-url-field">
          <span>Base URL</span>
          <input id="model-provider-url" type="url" value="http://localhost:1234" required spellcheck="false">
        </label>
        <label class="endpoint-key-field" id="endpoint-key-field" hidden>
          <span>API key</span>
          <input id="model-provider-key" type="password" autocomplete="off" placeholder="Optional local token">
        </label>
        <button class="button primary" id="add-model-provider" type="submit">Add &amp; test</button>
      </form>

      <div class="model-connect-status" id="model-connect-status" role="status" aria-live="polite"></div>
      <div class="model-provider-list" id="model-provider-list" aria-live="polite"></div>
    </section>

    <section class="folder-explainer" aria-labelledby="created-title">
      <div>
        <p class="eyebrow">What happens next</p>
        <h2 id="created-title">One root, separate app folders</h2>
      </div>
      <p>The apps create a small registry file and a separate folder for each project you use, such as <code>prompt-gallery</code> or <code>game-forge-v2</code>. Saving is append-only by default.</p>
    </section>

    <div class="settings-actions" aria-label="Open a project">
      <a class="button" href="./#projects">Browse all projects</a>
      <a class="button" href="game-forge-v2/">Open Game Forge</a>
      <a class="button" href="prompt-gallery/">Open Prompt Gallery</a>
    </div>
  </main>

  <footer><span>Peter Edwards · Projects MIT licensed</span><span><a href="privacy.html">Privacy</a></span></footer>
  <script defer src="https://static.cloudflareinsights.com/beacon.min.js" data-cf-beacon='{"token": "087e962bcbbc4be0bc1ba8bf21be0bde"}'></script>
</body>
</html>