llmgw/internal/web/templates/admin_mirror.html

{{define "admin_mirror"}}<!doctype html>
<html lang="en"><head>
<meta charset="utf-8"><meta name="viewport" content="width=device-width,initial-scale=1">
{{/* Only while something is downloading. A reload at any other time would
     take with it whatever the operator was typing into the form below. */}}
{{if .RefreshSec}}<meta http-equiv="refresh" content="{{.RefreshSec}}">{{end}}
<title>Mirror · llmgw</title><link rel="stylesheet" href="{{stylesheet}}">
</head><body><div class="wrap">
<header>
  <h1>Model mirror</h1>
  <p class="sub">
    {{if .Unreachable}}{{.Addr}} did not answer{{else}}{{.Present}} held ·
    {{.Fetching}} downloading · {{.Failed}} failed · {{.Addr}}{{end}}
  </p>
  {{template "nav" .Nav}}
</header>

{{if .Error}}<p class="error">{{.Error}}</p>{{end}}
{{if .Message}}<p class="ok">{{.Message}}</p>{{end}}

{{if .Unreachable}}
{{/* Said outright rather than shown as an empty table. A mirror holding
     nothing and a mirror that is not there look the same in a list of zero
     rows, and they call for opposite responses. */}}
<p class="note">Nothing was read from the mirror, so this page is not a list of
what it holds — it is not currently reachable at <code>{{.Addr}}</code>. Check
that <code>llmgw-mirror</code> is running there and that its management port is
reachable from this machine.</p>
{{else}}

<h2>What the mirror holds</h2>
{{if .Rows}}
<table class="compact held">
<tr><th>Model</th><th>Files</th><th class="num">Size</th><th>Runs</th><th></th></tr>
{{range .Groups}}{{$grouped := gt .Models 1}}
{{if $grouped}}
<tr class="model-group"><td colspan="5"><strong>{{.Family}}</strong>
  <span class="opt">{{.Models}} models · {{bytes .Bytes}}</span>
  {{if .Fetching}}<span class="tag new">downloading</span>{{end}}
  {{if .Failed}}<span class="tag bad">failed</span>{{end}}
  {{if .InUse}}<span class="tag">in use</span>{{end}}</td></tr>
{{end}}
{{range .Rows}}<tr{{if $grouped}} class="model-version"{{end}}>
  <td><strong>{{.Model}}</strong> <span class="opt">{{.Version}}</span></td>
  <td class="file" title="{{.File}}{{if .SHA256}} · sha256 {{.SHA256}}{{end}}">
    {{if gt .Files 1}}{{.Files}} files{{else}}<code>{{.File}}</code>{{end}}</td>
  <td class="num">
    {{if eq (printf "%s" .State) "fetching"}}
      {{if .HasPercent}}{{.Percent}}% of {{bytes .BytesTotal}}
      {{else}}{{/* No total came from upstream, so no percentage is invented —
                   a bar that guesses its own end and then stops moving is worse
                   than a number that only goes up. */}}
      {{bytes .BytesDone}} so far{{end}}
    {{else if eq (printf "%s" .State) "failed"}}—
    {{else}}{{bytes .BytesDone}}{{end}}
  </td>
  <td>
    {{if eq (printf "%s" .State) "fetching"}}<span class="state state-starting">downloading</span>
    {{else if eq (printf "%s" .State) "failed"}}<span class="state state-failed">failed</span>
    {{else if .UsedBy}}{{$row := .}}{{range .UsedBy}}<div class="use">
      {{/* Named only when the catalog calls it something other than the
           mirror does; otherwise it is the model column again. */}}
      {{if ne .Name $row.Model}}<span class="as" title="{{.Name}}">as {{.Name}}</span>{{end}}
      {{if .Declared}}<span class="opt">{{if eq .Mode "always"}}always on{{else if eq .Mode "ondemand"}}on demand{{else}}stopped{{end}}<br>set in server.yaml</span>
      {{else}}
      {{/* Submitted on change: the choice is the action, and a second click
           to confirm it would only be a way to forget to. */}}
      <form method="post" action="/admin/mirror/mode" class="inline">
        <input type="hidden" name="csrf" value="{{$.CSRF}}">
        <input type="hidden" name="model" value="{{.Name}}">
        <input type="hidden" name="replicas" value="{{.Replicas}}">
        <select name="mode" class="mode" onchange="this.form.submit()" aria-label="How {{.Name}} runs">
          <option value="always"{{if eq .Mode "always"}} selected{{end}}>always on</option>
          <option value="ondemand"{{if eq .Mode "ondemand"}} selected{{end}}>on demand</option>
          <option value="stopped"{{if eq .Mode "stopped"}} selected{{end}}>stopped</option>
        </select>
        <noscript><button type="submit" class="link">set</button></noscript>
      </form>
      {{end}}
    </div>{{end}}
    {{else}}<span class="opt">not served</span>{{end}}
  </td>
  <td class="actions">
    {{if eq (printf "%s" .State) "failed"}}
    {{/* Retry, not remove. A fetch that failed left nothing on disk, so the
         remove would come back "the mirror holds nothing for this" — a button
         whose only outcome is a confusing error. Starting the same download
         again is what somebody looking at a failed row wants, and the mirror
         allows it: only a download still running is refused. */}}
    <form method="post" action="/admin/mirror/fetch" class="inline">
      <input type="hidden" name="csrf" value="{{$.CSRF}}">
      <input type="hidden" name="source" value="{{.Source}}">
      <input type="hidden" name="model" value="{{.Model}}">
      <input type="hidden" name="version" value="{{.Version}}">
      <input type="hidden" name="file" value="{{.File}}">
      <button type="submit" class="link">retry</button>
    </form>
    {{else if eq (printf "%s" .State) "present"}}
    <button type="button" class="link copy-url" data-url="{{.URL}}" title="{{.URL}}">copy address</button>
    {{if and $.HasCatalog (not .UsedBy)}}{{if .Serve}}<a href="/admin/mirror">close</a>
    {{else}}<a href="/admin/mirror?serve={{.Model}}:{{.Version}}">serve</a>{{end}}{{end}}
    <form method="post" action="/admin/mirror/{{.Model}}/{{.Version}}/remove"
          class="inline" onsubmit="return confirm('Remove {{.Model}}/{{.Version}} and every file in it?{{if .UsedBy}} It is in use by the catalog.{{end}}')">
      <input type="hidden" name="csrf" value="{{$.CSRF}}">
      <button type="submit" class="link danger">remove</button>
    </form>
    {{end}}
  </td>
</tr>
{{if .Err}}<tr><td colspan="5" class="error-row">{{.Err}}</td></tr>{{end}}

{{with .Serve}}
<tr class="editing"><td colspan="5">
{{if .Refusal}}
{{/* Offered and then failing at placement is worse than not offered at all. */}}
<p class="error">Cannot be served as it stands: {{.Refusal}}.</p>
{{else}}
{{if .Warning}}<p class="note error-row">{{.Warning}}.</p>{{end}}
<form method="post" action="/admin/mirror/serve">
  <input type="hidden" name="csrf" value="{{$.CSRF}}">
  <input type="hidden" name="source" value="{{.Source}}">
  <input type="hidden" name="engine" value="{{.Engine}}">
  <p class="sub">Serving <code>{{.Source}}</code> with <strong>{{.Engine}}</strong>.
  {{if .Files}}The whole version, {{.Files}} files: the node fetches them into one
  directory and the engine is given the directory.{{end}}</p>
  <div class="field-row">
    <label>Name <input type="text" name="name" value="{{.Model}}" required></label>
    <label>Aliases <span class="opt">comma separated</span>
      <input type="text" name="aliases" placeholder="coder-7b"></label>
    <label>Replicas <input type="text" name="replicas" value="1" required></label>
    <label>Runs
      <select name="mode">
        <option value="always">always on</option>
        <option value="ondemand">on demand: loads when asked, stops after 15 min unused</option>
      </select></label>
  </div>
  {{/* Reaches clients through /v1/models, which for most of them is the only
       view of what this gateway serves. Nine names and nothing to choose
       between them is a choice made by guesswork. */}}
  <label>What it is for <span class="opt">shown to clients in /v1/models</span>
    <input type="text" name="description"
           placeholder="general chat, strong Swedish; 14B, slower than the 8B"></label>
  <div class="field-row">
    <label>Context a request may use
      <input type="text" name="context_length" value="4096" required></label>
    <label>Output limit <span class="opt">optional</span>
      <input type="text" name="max_output" value="1024"></label>
    <label>Engine context pool <span class="opt">divided across slots</span>
      <input type="text" name="ctx_size" value="16384"></label>
  </div>
  <div class="field-row">
    <label>Profile name <input type="text" name="profile" value="{{.Profile}}"></label>
    <label>Needs RAM (GB) <span class="opt">estimated from the weights</span>
      <input type="text" name="min_memory_gb" value="{{.MinMemoryGB}}"></label>
    <label>Only nodes labelled <span class="opt">key / value, optional</span>
      <input type="text" name="selector_key" placeholder="role" size="8">
      <input type="text" name="selector_value" placeholder="cpu-large" size="10"></label>
  </div>
  <button type="submit">Add to the catalog</button>
</form>
<p class="note">The context pool is what the engine is given; llama.cpp divides
it across its parallel slots, so the number one request may use is the smaller
one above.</p>
<p class="note">The memory figure is the weights plus half again for the KV pool
plus two for the engine — an estimate, and editable. It is filled in rather than
left blank because a CPU profile has no card to be short of: with no figure,
every ready node is a candidate and the first one picked is whichever the
scheduler reached, not whichever can hold the model. A node label does the same
job by name if you would rather pin it.</p>
{{end}}
</td></tr>
{{end}}
{{end}}
{{end}}
</table>
<p class="note">“copy address” gives what a catalog entry's <code>source</code>
should say — the file, or for a version of several files the directory, with its
trailing slash. It is rendered rather than assembled by hand because the mirror
and the catalog have to spell a file the same way, which is also how this page
knows which models use it.</p>
<script>
document.addEventListener('click', function (e) {
  var b = e.target.closest('.copy-url');
  if (!b) return;
  var done = function (t) { var o = b.textContent; b.textContent = t; setTimeout(function () { b.textContent = o; }, 1500); };
  if (navigator.clipboard) {
    navigator.clipboard.writeText(b.dataset.url).then(function () { done('copied'); },
      function () { window.prompt('Address', b.dataset.url); });
  } else { window.prompt('Address', b.dataset.url); }
});
</script>
{{else}}<p class="empty">The mirror is reachable and holds nothing yet.</p>{{end}}

{{if .HasRegistry}}
<h2>Explore models</h2>
{{if not .Repo}}{{with .Explore}}
<nav class="pills" aria-label="Which models">{{range .Views}}<a href="{{.Href}}"{{if .Active}} class="on" aria-current="page"{{end}}>{{.Label}}</a>{{end}}</nav>
<div class="explore-filters">
  <nav class="pills small" aria-label="Task">{{range .Tasks}}<a href="{{.Href}}"{{if .Active}} class="on" aria-current="true"{{end}}>{{.Label}}</a>{{end}}</nav>
  <nav class="pills small" aria-label="Size">{{range .Sizes}}<a href="{{.Href}}"{{if .Active}} class="on" aria-current="true"{{end}}>{{.Label}}</a>{{end}}</nav>
</div>
{{end}}
<form method="get" action="/admin/mirror" class="explore-search">
  {{with .Explore}}<input type="hidden" name="view" value="{{.View}}">
  <input type="hidden" name="task" value="{{.Task}}">
  <input type="hidden" name="size" value="{{.Size}}">
  <input type="hidden" name="x" value="1">{{end}}
  <input type="search" name="q" value="{{.Query}}" placeholder="Search Hugging Face — qwen coder, gemma, bge…" aria-label="Search Hugging Face">
  <label class="check"><input type="checkbox" name="here" value="1" onchange="this.form.submit()"{{if .Explore}}{{if .Explore.Here}} checked{{end}}{{end}}> only what these nodes can run</label>
  <button type="submit">Search</button>
</form>
{{end}}

{{if .RegistryErr}}
{{/* Its own line, not the page's error: the registry is somebody else's server
     and it being down says nothing about the mirror, which is what this page is
     for and has to keep working when Hugging Face does not. */}}
<p class="note error-row">The registry: {{.RegistryErr}}</p>
{{end}}

{{if not .Repo}}{{with .Explore}}
{{if .Rows}}
<table class="explore">
<tr><th>Repository</th><th class="num">Size</th><th>Architecture</th><th class="num">Context</th>
<th class="num">Downloads</th><th class="num">Likes</th><th class="num">Updated</th></tr>
{{range .Rows}}<tr class="explore-row">
  <td><a href="{{.Href}}" class="open-repo" data-repo="{{.ID}}" aria-expanded="false"><code>{{.ID}}</code></a>
    {{if .New}}<span class="tag new">new</span>{{end}}{{if .Gated.Is}}<span class="tag">gated</span>{{end}}
    {{if or .BaseModel .License}}<div class="from">{{with .BaseModel}}from {{.}}{{end}}{{if and .BaseModel .License}} · {{end}}{{.License}}</div>{{end}}</td>
  <td class="num">{{.ParamsLabel}}</td>
  <td>{{if .Architecture}}<span{{if not .Verdict.Servable}} class="cannot"{{end}} title="{{.Verdict.Reason}}">{{.Architecture}}{{if .Verdict.Checked}}{{if .Verdict.Servable}} ✓{{else}} ✗{{end}}{{end}}</span>{{else}}—{{end}}</td>
  <td class="num">{{.ContextLabel}}</td>
  <td class="num">{{compact .Downloads}}</td>
  <td class="num">{{compact .Likes}}</td>
  <td class="num">{{.Age}}</td>
</tr>{{end}}
</table>
<script>
// A repository opens beneath its row rather than replacing the list: the
// list is what somebody is choosing from, and losing it to read one card
// means finding their place again for the next. The link still goes to the
// repository's own page when this does not run, or with a modifier key.
document.querySelector("table.explore").addEventListener("click", async function (e) {
  const a = e.target.closest("a.open-repo");
  if (!a || e.metaKey || e.ctrlKey || e.shiftKey || e.button !== 0) return;
  e.preventDefault();
  const row = a.closest("tr");
  let detail = row.nextElementSibling;
  if (detail && detail.classList.contains("repo-detail")) {
    detail.hidden = !detail.hidden;
    a.setAttribute("aria-expanded", String(!detail.hidden));
    row.classList.toggle("open", !detail.hidden);
    return;
  }
  detail = document.createElement("tr");
  detail.className = "repo-detail";
  const td = document.createElement("td");
  td.colSpan = row.children.length;
  td.textContent = "Loading…";
  detail.appendChild(td);
  row.after(detail);
  row.classList.add("open");
  a.setAttribute("aria-expanded", "true");
  try {
    const resp = await fetch("/admin/mirror/repo?repo=" + encodeURIComponent(a.dataset.repo));
    const text = await resp.text();
    if (!resp.ok) { td.textContent = text.slice(0, 300); td.classList.add("error-row"); return; }
    // Our own server's rendering of the repository, the card already
    // stripped of anything that runs: the same HTML the full page shows.
    td.innerHTML = text;
  } catch (err) {
    td.textContent = "Could not load it: " + err;
  }
});
</script>
{{if .Hidden}}<p class="note">{{.Hidden}} more {{if eq .Hidden 1}}is{{else}}are{{end}} hidden: their architecture is one the nodes' llama.cpp cannot load.
Untick “only what these nodes can run” to see them.</p>{{end}}
<p class="note">A gated repository needs its terms accepted with a Hugging Face
account, and this browses anonymously — the files will list, and fetching one
will fail at the mirror. Download it by hand and put it in the mirror
directory, or give the mirror a token.</p>
{{else if not $.RegistryErr}}
<p class="empty">{{if $.Query}}Nothing matched “{{$.Query}}”{{else}}Nothing to list{{end}}{{if .Hidden}} that these nodes can run — {{.Hidden}} {{if eq .Hidden 1}}result was{{else}}results were{{end}} hidden for an architecture llama.cpp cannot load{{end}}.</p>
{{end}}
{{end}}{{end}}

{{if .Repo}}{{template "mirror_repo" .}}{{end}}
{{end}}

<h2>Fetch a model</h2>
<p class="note">By hand, for anything the search does not reach.</p>
<form method="post" action="/admin/mirror/fetch">
  <input type="hidden" name="csrf" value="{{.CSRF}}">
  <label>Source URL
    <input type="text" name="source" required
           placeholder="https://huggingface.co/…/resolve/main/model.gguf"></label>
  <div class="field-row">
    <label>Model <input type="text" name="model" required placeholder="qwen2.5-coder-7b"></label>
    <label>Version <input type="text" name="version" required placeholder="2026-09"></label>
    <label>File <span class="opt">optional</span>
      <input type="text" name="file" placeholder="taken from the URL"></label>
  </div>
  <label class="inline-check"><input type="checkbox" name="force" value="1">
    Fetch anyway <span class="opt">even if the nodes' llama.cpp cannot load it</span></label>
  <button type="submit">Start download</button>
</form>
<p class="note">The mirror answers as soon as the download is under way, not when
it finishes — a model is tens of gigabytes and nothing here holds a connection
open for that. The row above will show it arriving. A version is never
overwritten: fetching into one that exists is refused, so a new build is a new
version and the old weights stay where they are.</p>
{{end}}

<footer><span>llmgw {{.Version}}</span><span>{{.Now}}</span></footer>
</div></body></html>{{end}}

{{/* One repository: its card and its files. The whole page when opened by
     address, and the row beneath a listing entry when opened from the list. */}}
{{define "mirror_repo"}}
{{if not .Inline}}<p><a href="{{.Back}}">← back to the list</a></p>
<h3>{{.Repo.ID}} <span class="opt">at {{.Repo.ShortRevision}}</span></h3>{{end}}
<p class="repo-meta">
  {{if .Inline}}<span>at <code>{{.Repo.ShortRevision}}</code></span>{{end}}
  {{with .Repo.BaseModel}}<span>from {{range $i, $b := .}}{{if $i}}, {{end}}<code>{{$b}}</code>{{end}}</span>{{end}}
  {{with .Repo.License}}<span>license {{.}}</span>{{end}}
  {{with .Repo.Languages}}<span>{{join .}}</span>{{end}}
  <span><a href="{{.RepoURL}}" rel="noopener noreferrer">on Hugging Face ↗</a></span>
</p>
{{if .Card}}
<details class="model-card" open>
  <summary>Model card</summary>
  <div class="card-body">{{.Card}}</div>
</details>
{{end}}
{{if .RepoSets}}
<table class="repo-files">
<tr><th>What</th><th class="num">Size</th><th>Published checksum</th><th></th></tr>
{{range .RepoSets}}<tr>
  <td><code>{{.Name}}</code>
    {{if .Shards}}<span class="opt">{{.Shards}} shards, fetched together</span>{{end}}</td>
  <td class="num">{{bytes .Bytes}}</td>
  <td>{{with .SHA256}}<code title="{{.}}">{{slice . 0 12}}</code>{{else}}—{{end}}</td>
  <td>
    {{if not .Verdict.Servable}}
    <span class="state state-failed" title="{{.Verdict.Reason}}">cannot be served</span>
    <div class="note">{{.Verdict.Reason}}</div>
    {{else}}
    {{if .Verdict.Architecture}}<span class="opt" title="{{.Verdict.Reason}}">{{.Verdict.Architecture}}{{if .Verdict.Checked}} ✓{{end}}</span>{{end}}
    <form method="post" action="/admin/mirror/fetch" class="inline">
      <input type="hidden" name="csrf" value="{{$.CSRF}}">
      {{range .Sources}}<input type="hidden" name="source" value="{{.URL}}">
      <input type="hidden" name="file" value="{{.Path}}">
      {{end}}
      <input type="text" name="model" value="{{.Model}}" size="24">
      <input type="text" name="version" value="{{.Version}}" size="10">
      <button type="submit">Fetch</button>
    </form>
    {{end}}
  </td>
</tr>{{end}}
</table>
<p class="note">Only the weights are listed; a repository is mostly a README and
a tokenizer. Shards of one file are one row and one button — a fifteen-shard
model is one decision, not fifteen. Two separate files are two rows because they
are usually alternatives: one quantisation or the other, not both.</p>
<p class="note">The source URLs are pinned to
<code>{{.Repo.ShortRevision}}</code> rather than to a branch, so what arrives is
what is listed here even if the branch moves in between — and a single file's
checksum above is what the registry publishes, to compare with what the mirror
computes once it has it.</p>
{{else}}
<p class="empty">No weight files in {{.Repo.ID}}.</p>
{{end}}
{{end}}