browser-based-LLM/index.html at main · harisnae/browser-based-LLM · 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
<!DOCTYPE html>
<html lang="en">
<head>
  <meta charset="UTF-8" />
  <meta name="viewport"
        content="width=device-width, initial-scale=1.0, maximum-scale=1.0, user-scalable=no" />
  <title>TinyLlama Browser Chat</title>
  <link rel="stylesheet" href="style.css">
  <meta name="description"
        content="Run TinyLlama 1.1B entirely in your browser with no server required">
</head>
<body>
  <div class="container">
    <header>
      <h1>💬 TinyLlama Browser Chat</h1>
      <div class="status">
        Model: TinyLlama-1.1B-Chat-v1.0 • Running entirely in your browser
      </div>
    </header>

    <main>
      <div class="chat-container">
        <div id="chat-history" class="chat-history"></div>

        <div class="input-area">
          <textarea id="input"
                    placeholder="Ask me anything..."
                    rows="3"
                    autocomplete="off"
                    autocorrect="off"
                    autocapitalize="off"
                    spellcheck="false"></textarea>

          <div class="controls">
            <button id="generate" class="primary">Generate</button>
            <button id="cancel" class="danger" style="display:none;">Cancel</button>
            <button id="clear" class="secondary">Clear</button>

            <div class="settings">
              <label>
                Max tokens:
                <input type="number" id="maxTokens" value="200" min="50" max="500">
              </label>
              <span class="info-tooltip"
                    title="Higher values allow longer responses but take more time and memory">ⓘ</span>
            </div>
          </div>
        </div>
      </div>

      <div class="info-panel">
        <div id="output" class="output">
          ✅ Ready! Click "Generate" to load the model and start chatting.<br><br>
          <strong>First-time loading note:</strong> The model is ~2 GB and may take 30-60 s to download.
          Subsequent visits are faster thanks to browser caching.
        </div>

        <div class="model-info">
          <p>This demo runs entirely in the browser using the
            <code>@xenova/transformers</code> library. No data leaves your device.</p>
          <p>Developed as a lightweight, privacy-focused demonstration of
            browser-based LLMs.</p>
        </div>
      </div>
    </main>

    <footer>
      <p>Made with ❤️ for GitHub Pages •
        <a href="https://huggingface.co/Xenova/TinyLlama-1.1B-Chat-v1.0"
           target="_blank">Model Info</a></p>
    </footer>
  </div>

  <!-- -------------------------------------------------
       Load the JavaScript that contains all the logic.
       defer guarantees the script runs after the DOM is built.
       ------------------------------------------------- -->
  <script src="script.js" type="module"></script>
</body>
</html>