runtime-tools/docs/index.html at main · Firefox-AI/runtime-tools · 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
<!DOCTYPE html>
<html lang="en">
<head>
  <meta charset="UTF-8">
  <meta name="viewport" content="width=device-width, initial-scale=1.0">
  <title>Firefox AI Runtime</title>
  <link rel="stylesheet" href="style.css">
  <style>
    .collection-name { font-family: monospace; font-size: 0.85em; color: #555; margin-bottom: 0; }
  </style>
</head>
<body class="main-body">

  <div class="main-header">
    <div>
      <div class="main-header-left">
        <h1>
          <img src="smart-window.svg" class="main-header-icon">Firefox AI Runtime
        </h1>
        <h2>Live views of the Remote Settings collections powering the Firefox AI runtime and Smart Window</h2>
      </div>
      <div class="main-header-right">
        <img class="main-header-noodles" src="noodles.png" aria-role="presentational">
      </div>
    </div>
  </div>

  <div class="main-links">
    <a href="./ml-inference-options">
      <h3>Inference Options</h3>
      <p>Which ML model is deployed for each Firefox feature — task type, model ID, quantization, and revision.</p>
      <p class="collection-name">ml-inference-options</p>
    </a>
    <a href="./ml-onnx-runtime">
      <h3>ONNX Runtime Binaries</h3>
      <p>The WebAssembly runtime files (ONNX and wllama) hosted in Remote Settings for on-device inference.</p>
      <p class="collection-name">ml-onnx-runtime</p>
    </a>
    <a href="./ml-model-allow-deny-list">
      <h3>Model Allow/Deny List</h3>
      <p>URL prefixes that are allowed or denied when Firefox loads models from external hosts.</p>
      <p class="collection-name">ml-model-allow-deny-list</p>
    </a>
    <a href="./ml-inference-words-block-list">
      <h3>Words Block List</h3>
      <p>Language-specific n-gram block lists used for content filtering during ML inference.</p>
      <p class="collection-name">ml-inference-words-block-list</p>
    </a>
    <a href="./ai-window-prompts">
      <h3>AI Window Prompts</h3>
      <p>System prompts and configuration for each Firefox Smart Window feature and AI model.</p>
      <p class="collection-name">ai-window-prompts</p>
    </a>
    <a href="./ai-window-tools">
      <h3>AI Window Tools</h3>
      <p>Tool/function definitions the Smart Window assistant can call — tab management, memories, search, and more.</p>
      <p class="collection-name">ai-window-tools</p>
    </a>
    <a href="./ai-window-telemetry-prompts">
      <h3>AI Window Telemetry Prompts</h3>
      <p>LLM prompts that classify conversations to produce telemetry, with their allowed output values.</p>
      <p class="collection-name">ai-window-telemetry-prompts</p>
    </a>
    <a href="./ai-window-telemetry-triggers">
      <h3>AI Window Telemetry Triggers</h3>
      <p>Sampling rules that decide when telemetry prompts run.</p>
      <p class="collection-name">ai-window-telemetry-triggers</p>
    </a>
    <a href="./ml-inference-request-options">
      <h3>Inference Request Options</h3>
      <p>Per-feature inference request options (system prompt, input size) stored as a JSON blob.</p>
      <p class="collection-name">ml-inference-request-options</p>
    </a>
  </div>

</body>
</html>