Feinschliff Phase 2: Dashboard Redesign, RAM Check, Accordions

This commit is contained in:
Hitonabi
2026-06-20 22:30:40 +02:00
parent e3be7fbfb5
commit 8b76adc96e
10 changed files with 358 additions and 119 deletions
+25
View File
@@ -43,6 +43,11 @@ class ChatReq(BaseModel):
message: str message: str
class UpdateReq(BaseModel):
alias: str
ctx: int
# --------------------------------------------------------------------------- # ---------------------------------------------------------------------------
# Endpoints # Endpoints
# --------------------------------------------------------------------------- # ---------------------------------------------------------------------------
@@ -144,6 +149,26 @@ def register(req: RegisterReq):
"note": "In config.yaml geschrieben. llama-swap mit -watch-config laedt automatisch neu."} "note": "In config.yaml geschrieben. llama-swap mit -watch-config laedt automatisch neu."}
@router.post("/update_model")
def update_model(req: UpdateReq):
cfg = read_config()
if req.alias not in cfg.get("models", {}):
raise HTTPException(404, "Modell nicht gefunden")
spec = cfg["models"][req.alias]
cmd = str(spec.get("cmd", ""))
# Replace or add context size
if re.search(r'-(?:c|-ctx-size)\s+\d+', cmd):
cmd = re.sub(r'-(?:c|-ctx-size)\s+\d+', f'-c {req.ctx}', cmd)
else:
cmd = cmd.strip() + f" -c {req.ctx}\n"
cfg["models"][req.alias]["cmd"] = LiteralScalarString(cmd)
write_config(cfg)
return {"ok": True}
@router.post("/unload") @router.post("/unload")
def unload(model: str | None = None): def unload(model: str | None = None):
path = f"/api/models/unload/{model}" if model else "/api/models/unload" path = f"/api/models/unload/{model}" if model else "/api/models/unload"
+17
View File
@@ -140,3 +140,20 @@ button:disabled{opacity:.5;cursor:not-allowed}
opacity:0;transition:.25s;pointer-events:none;max-width:90vw;z-index:50} opacity:0;transition:.25s;pointer-events:none;max-width:90vw;z-index:50}
.toast.show{opacity:1} .toast.show{opacity:1}
.toast.err{border-color:var(--err);color:#ffb4ae} .toast.err{border-color:var(--err);color:#ffb4ae}
/* ---- Accordion (Guides) ---- */
.guide-acc { border-bottom: 1px solid var(--line); }
.guide-acc:last-child { border-bottom: none; }
.guide-acc summary {
padding: 16px 20px;
font-weight: 600;
cursor: pointer;
list-style: none;
display: flex;
justify-content: space-between;
align-items: center;
}
.guide-acc summary::-webkit-details-marker { display: none; }
.guide-acc summary::after { content: "▼"; font-size: 10px; color: var(--mut); transition: transform 0.2s; }
.guide-acc[open] summary::after { transform: rotate(180deg); }
.guide-acc .acc-body { padding: 0 20px 20px 20px; }
+23 -7
View File
@@ -22,7 +22,7 @@
<a class="nav-item" data-view="guides" title="Guides" data-ic="help"></a> <a class="nav-item" data-view="guides" title="Guides" data-ic="help"></a>
</nav> </nav>
<div class="side-foot"> <div class="side-foot">
<span class="nav-item disabled" title="Einstellungen (bald)" data-ic="settings"></span> <span class="nav-item" title="Einstellungen" data-ic="settings" id="nav-settings"></span>
</div> </div>
</aside> </aside>
@@ -31,11 +31,7 @@
<header class="topbar"> <header class="topbar">
<span class="status-pill"><span id="swdot" class="dot"></span><span id="swlabel">verbinde…</span></span> <span class="status-pill"><span id="swdot" class="dot"></span><span id="swlabel">verbinde…</span></span>
<span class="spacer"></span> <span class="spacer"></span>
<span class="top-stat">Aktiv:<b id="top-active" style="color:var(--teal)"></b></span> <span class="top-stat" id="top-active-text" style="color:var(--teal)">Kein Modell geladen</span>
<span class="top-stat" style="margin-left:8px">Modelle<b id="top-models"></b></span>
<span class="top-stat">Jobs<b id="top-jobs">0</b></span>
<span class="top-clock" id="clock">--:--</span>
<input id="token" class="tokin" placeholder="Token" autocomplete="off">
</header> </header>
<!-- Alert-Banner (wird per JS ein-/ausgeblendet) --> <!-- Alert-Banner (wird per JS ein-/ausgeblendet) -->
@@ -46,7 +42,13 @@
<section class="view" data-view="overview"> <section class="view" data-view="overview">
<div id="hero"></div> <div id="hero"></div>
<div class="card" id="ov-models" style="margin-top: 24px"></div>
<div class="grid grid-3" id="ov-quick" style="margin-top: 24px"></div>
<div class="grid grid-2" style="margin-top: 24px; align-items: start;">
<div class="card" id="ov-models"></div>
<div class="card" id="ov-recent-jobs"></div>
</div>
</section> </section>
<section class="view" data-view="activity" hidden> <section class="view" data-view="activity" hidden>
@@ -78,6 +80,20 @@
</div> </div>
</div> </div>
<!-- Settings Modal -->
<div id="settings-modal" style="display:none; position:fixed; top:0; left:0; right:0; bottom:0; background:rgba(0,0,0,0.8); z-index:100; align-items:center; justify-content:center;">
<div class="card" style="width:100%; max-width:400px; position:relative">
<button id="sm-close" class="ghost" style="position:absolute; top:12px; right:12px;">Schließen</button>
<h2 style="margin-top:0">Einstellungen</h2>
<div style="margin-top:24px">
<label>API Token (Authentifizierung)</label>
<input id="token" class="tokin" placeholder="Optionales API Token..." autocomplete="off">
<div class="meta" style="font-size:12px; margin-top:4px;">Wird für die WebSockets und API Calls genutzt, falls der Server geschützt ist.</div>
</div>
</div>
</div>
<div id="toast" class="toast"></div> <div id="toast" class="toast"></div>
<!-- Icons in die Nav/Logo einsetzen, bevor das Haupt-Modul laedt --> <!-- Icons in die Nav/Logo einsetzen, bevor das Haupt-Modul laedt -->
+1
View File
@@ -55,6 +55,7 @@ export const ICON = {
alert: _svg('<path d="M10.3 3.9 1.8 18a2 2 0 0 0 1.7 3h17a2 2 0 0 0 1.7-3L13.7 3.9a2 2 0 0 0-3.4 0z"/><path d="M12 9v4M12 17h.01"/>'), alert: _svg('<path d="M10.3 3.9 1.8 18a2 2 0 0 0 1.7 3h17a2 2 0 0 0 1.7-3L13.7 3.9a2 2 0 0 0-3.4 0z"/><path d="M12 9v4M12 17h.01"/>'),
gauge: _svg('<path d="M12 14 16 9"/><circle cx="12" cy="13" r="9"/><path d="M12 4v2M21 13h-2M5 13H3"/>'), gauge: _svg('<path d="M12 14 16 9"/><circle cx="12" cy="13" r="9"/><path d="M12 4v2M21 13h-2M5 13H3"/>'),
search: _svg('<circle cx="11" cy="11" r="8"/><path d="m21 21-4.3-4.3"/>'), search: _svg('<circle cx="11" cy="11" r="8"/><path d="m21 21-4.3-4.3"/>'),
compass: _svg('<circle cx="12" cy="12" r="10"/><polygon points="16.24 7.76 14.12 14.12 7.76 16.24 9.88 9.88 16.24 7.76"/>'),
code: _svg('<polyline points="16 18 22 12 16 6"/><polyline points="8 6 2 12 8 18"/>'), code: _svg('<polyline points="16 18 22 12 16 6"/><polyline points="8 6 2 12 8 18"/>'),
eye: _svg('<path d="M2 12s3-7 10-7 10 7 10 7-3 7-10 7-10-7-10-7Z"/><circle cx="12" cy="12" r="3"/>'), eye: _svg('<path d="M2 12s3-7 10-7 10 7 10 7-3 7-10 7-10-7-10-7Z"/><circle cx="12" cy="12" r="3"/>'),
}; };
+17 -6
View File
@@ -24,17 +24,19 @@ function applyStatus(s) {
if (!s) { if (!s) {
dot.className = "dot off"; dot.className = "dot off";
label.textContent = "Backend nicht erreichbar"; label.textContent = "Backend nicht erreichbar";
$("#top-models").textContent = ""; $("#top-active-text").textContent = "Backend offline";
$("#top-active").textContent = "";
showAlert("Backend nicht erreichbar läuft uvicorn?", false); showAlert("Backend nicht erreichbar läuft uvicorn?", false);
} else { } else {
const host = s.swap_url.replace(/^https?:\/\//, ""); const host = s.swap_url.replace(/^https?:\/\//, "");
dot.className = "dot " + (s.swap_ok ? "on" : "off"); dot.className = "dot " + (s.swap_ok ? "on" : "off");
label.textContent = (s.swap_ok ? "LLM-Engine: Online " : "LLM-Engine: Offline ") + host; label.textContent = (s.swap_ok ? "LLM-Engine: Online " : "LLM-Engine: Offline ") + host;
$("#top-models").textContent = (s.models || []).length;
const active = (s.models || []).find(m => m.state === "running"); const active = (s.models || []).find(m => m.state === "running");
$("#top-active").textContent = active ? active.name : ""; if (active) {
$("#top-active-text").innerHTML = `Geladen: <b style="color:var(--teal)">${active.name}</b>`;
} else {
$("#top-active-text").innerHTML = "Kein Modell im VRAM";
}
if (s.swap_ok) hideAlert(); if (s.swap_ok) hideAlert();
else showAlert(`LLM-Engine nicht erreichbar unter <b>${host}</b> läuft der llama-swap Dienst?`, true); else showAlert(`LLM-Engine nicht erreichbar unter <b>${host}</b> läuft der llama-swap Dienst?`, true);
@@ -44,7 +46,6 @@ function applyStatus(s) {
function applyJobs(jobs) { function applyJobs(jobs) {
lastJobs = jobs || []; lastJobs = jobs || [];
$("#top-jobs").textContent = lastJobs.filter(j => j.state === "running" || j.state === "queued").length;
for (const p of panels) p.onJobs?.(lastJobs); for (const p of panels) p.onJobs?.(lastJobs);
} }
@@ -79,9 +80,19 @@ async function pollSystem() {
// ---- Boot ---- // ---- Boot ----
function bootToken() { function bootToken() {
const i = $("#token"); const i = $("#token");
if (i) {
i.value = getToken(); i.value = getToken();
i.addEventListener("change", e => { setToken(e.target.value); pollStatus(); }); i.addEventListener("change", e => { setToken(e.target.value); pollStatus(); });
} }
const sbtn = $("#nav-settings");
const smod = $("#settings-modal");
const scls = $("#sm-close");
if (sbtn && smod && scls) {
sbtn.addEventListener("click", () => smod.style.display = "flex");
scls.addEventListener("click", () => smod.style.display = "none");
}
}
function tickClock() { function tickClock() {
$("#clock").textContent = new Date().toTimeString().slice(0, 5); $("#clock").textContent = new Date().toTimeString().slice(0, 5);
} }
@@ -98,4 +109,4 @@ pollSystem();
setInterval(tickClock, 1000); setInterval(tickClock, 1000);
setInterval(pollStatus, 3000); setInterval(pollStatus, 3000);
setInterval(pollJobs, 1500); setInterval(pollJobs, 1500);
setInterval(pollSystem, 3000); setInterval(pollSystem, 2000);
+67 -19
View File
@@ -104,6 +104,14 @@ function mount() {
</div> </div>
</div> </div>
<div id="cb-m-fit-container" style="margin-top:24px; padding:16px; background:var(--bg); border:1px solid var(--line); border-radius:8px; display:flex; justify-content:space-between; align-items:center;">
<div>
<div style="font-weight:600; font-size:14px; margin-bottom:4px;">Ressourcen-Check</div>
<div class="meta" id="cb-m-fit-text">Berechne...</div>
</div>
<div id="cb-m-fit-badge"></div>
</div>
<button class="primary" id="cb-m-download" style="width:100%; margin-top:24px; padding:12px">Herunterladen & Einpflegen</button> <button class="primary" id="cb-m-download" style="width:100%; margin-top:24px; padding:12px">Herunterladen & Einpflegen</button>
</div> </div>
</div> </div>
@@ -114,9 +122,46 @@ function mount() {
$("#cb-modal-close").addEventListener("click", () => $("#cb-modal").style.display = "none"); $("#cb-modal-close").addEventListener("click", () => $("#cb-modal").style.display = "none");
$("#cb-m-download").addEventListener("click", doDownload); $("#cb-m-download").addEventListener("click", doDownload);
$("#cb-m-files").addEventListener("change", updateLiveFit);
$("#cb-m-ctx").addEventListener("input", updateLiveFit);
renderCurated(); renderCurated();
} }
function extractParamsB(name) {
const moe = name.match(/(\d+)x(\d+(?:\.\d+)?)[bB]/i);
if (moe) return parseInt(moe[1]) * parseFloat(moe[2]);
const m = name.match(/(\d+(?:\.\d+)?)[bB](?![a-zA-Z])/i);
if (m) return parseFloat(m[1]);
return 7; // Fallback
}
function extractQuant(filename) {
const m = filename.match(/(Q\d_[A-Z0-9_]+|IQ\d_[A-Z0-9_]+|FP16|BF16)/i);
return m ? m[1].toUpperCase() : "Q4_K_M";
}
function updateLiveFit() {
const repo = $("#cb-m-repo").textContent;
const file = $("#cb-m-files").value;
const ctx = parseInt($("#cb-m-ctx").value) || 8192;
if (!repo || !file) {
$("#cb-m-fit-container").style.display = "none";
return;
}
const params_b = extractParamsB(repo);
const quant = extractQuant(file);
const m = { params_b, quant, ctx };
const fit = getFit(m, lastSys);
$("#cb-m-fit-container").style.display = "flex";
$("#cb-m-fit-text").innerHTML = `Geschätzter Bedarf: <b>~${fit.req.toFixed(1)} GB RAM/VRAM</b> <br><small class="meta">${params_b}B Params · ${quant}</small>`;
$("#cb-m-fit-badge").innerHTML = `<span class="badge ${fit.class}">${fit.text}</span>`;
}
async function doSearch() { async function doSearch() {
const q = $("#cb-search").value.trim(); const q = $("#cb-search").value.trim();
if (!q) return renderCurated(); if (!q) return renderCurated();
@@ -127,13 +172,13 @@ async function doSearch() {
$("#cb-grid").innerHTML = `<div class="meta" style="grid-column:1/-1; text-align:center; padding:40px">Suche auf HuggingFace...</div>`; $("#cb-grid").innerHTML = `<div class="meta" style="grid-column:1/-1; text-align:center; padding:40px">Suche auf HuggingFace...</div>`;
try { try {
const url = \`https://huggingface.co/api/models?search=\${encodeURIComponent(q)}&filter=gguf&sort=downloads&direction=-1&limit=12\`; const url = `https://huggingface.co/api/models?search=${encodeURIComponent(q)}&filter=gguf&sort=downloads&direction=-1&limit=12`;
const r = await fetch(url); const r = await fetch(url);
const data = await r.json(); const data = await r.json();
currentResults = data; currentResults = data;
renderResults(data); renderResults(data);
} catch (e) { } catch (e) {
$("#cb-grid").innerHTML = \`<div class="alert err" style="grid-column:1/-1">\${esc(e.message)}</div>\`; $("#cb-grid").innerHTML = `<div class="alert err" style="grid-column:1/-1">${esc(e.message)}</div>`;
} }
btn.disabled = false; btn.textContent = "Suchen"; btn.disabled = false; btn.textContent = "Suchen";
} }
@@ -145,17 +190,17 @@ function renderResults(results) {
return; return;
} }
grid.innerHTML = results.map((m, i) => \` grid.innerHTML = results.map((m, i) => `
<div class="card" style="display:flex; flex-direction:column; cursor:pointer" onclick="window.openModelModal(\${i})"> <div class="card" style="display:flex; flex-direction:column; cursor:pointer" onclick="window.openModelModal(${i})">
<h3 style="margin:0; font-size:16px; word-break:break-all">\${esc(m.id.split('/').pop())}</h3> <h3 style="margin:0; font-size:16px; word-break:break-all">${esc(m.id.split('/').pop())}</h3>
<div class="meta" style="font-size:12px; margin-top:4px">\${esc(m.author)}</div> <div class="meta" style="font-size:12px; margin-top:4px">${esc(m.author)}</div>
<div style="flex:1; margin-top:16px;"></div> <div style="flex:1; margin-top:16px;"></div>
<div style="display:flex; justify-content:space-between; align-items:center; margin-top:16px"> <div style="display:flex; justify-content:space-between; align-items:center; margin-top:16px">
<span class="badge b-idle" style="font-size:11px">GGUF</span> <span class="badge b-idle" style="font-size:11px">GGUF</span>
<span class="meta" style="font-size:12px">⬇ \${(m.downloads||0).toLocaleString()}</span> <span class="meta" style="font-size:12px">⬇ ${(m.downloads||0).toLocaleString()}</span>
</div> </div>
</div> </div>
\`).join(""); `).join("");
} }
// Global hook for inline onclick // Global hook for inline onclick
@@ -172,7 +217,7 @@ window.openModelModal = async (index) => {
$("#cb-m-loading").style.display = "block"; $("#cb-m-loading").style.display = "block";
try { try {
const url = \`https://huggingface.co/api/models/\${m.id}/tree/main\`; const url = `https://huggingface.co/api/models/${m.id}/tree/main`;
const r = await fetch(url); const r = await fetch(url);
const tree = await r.json(); const tree = await r.json();
const files = tree.filter(f => f.path.endsWith('.gguf')).map(f => f.path); const files = tree.filter(f => f.path.endsWith('.gguf')).map(f => f.path);
@@ -183,9 +228,11 @@ window.openModelModal = async (index) => {
if (files.length === 0) { if (files.length === 0) {
$("#cb-m-files").innerHTML = "<option value=''>Keine GGUF-Dateien im Hauptverzeichnis gefunden.</option>"; $("#cb-m-files").innerHTML = "<option value=''>Keine GGUF-Dateien im Hauptverzeichnis gefunden.</option>";
$("#cb-m-download").disabled = true; $("#cb-m-download").disabled = true;
$("#cb-m-fit-container").style.display = "none";
} else { } else {
$("#cb-m-files").innerHTML = files.map(f => \`<option value="\${esc(f)}">\${esc(f)}</option>\`).join(""); $("#cb-m-files").innerHTML = files.map(f => `<option value="${esc(f)}">${esc(f)}</option>`).join("");
$("#cb-m-download").disabled = false; $("#cb-m-download").disabled = false;
updateLiveFit();
} }
} catch(e) { } catch(e) {
$("#cb-m-loading").textContent = "Fehler beim Laden der Dateien."; $("#cb-m-loading").textContent = "Fehler beim Laden der Dateien.";
@@ -235,21 +282,21 @@ function renderCurated() {
if (!grid) return; if (!grid) return;
grid.innerHTML = CURATED_MODELS.map((m, i) => { grid.innerHTML = CURATED_MODELS.map((m, i) => {
const fit = getFit(m, lastSys); const fit = getFit(m, lastSys);
return \` return `
<div class="card" style="display:flex; flex-direction:column; cursor:pointer" onclick="window.openCuratedModal(\${i})"> <div class="card" style="display:flex; flex-direction:column; cursor:pointer" onclick="window.openCuratedModal(${i})">
<div style="display:flex; justify-content:space-between; align-items:center;"> <div style="display:flex; justify-content:space-between; align-items:center;">
<h3 style="margin:0; font-size:16px">\${esc(m.name)}</h3> <h3 style="margin:0; font-size:16px">${esc(m.name)}</h3>
<span class="badge \${fit.class}">\${fit.text}</span> <span class="badge ${fit.class}">${fit.text}</span>
</div> </div>
<div style="font-size:13px; color:var(--mut); margin-top:12px; flex:1; line-height:1.5;"> <div style="font-size:13px; color:var(--mut); margin-top:12px; flex:1; line-height:1.5;">
\${m.desc} ${m.desc}
</div> </div>
<div style="display:flex; justify-content:space-between; margin-top:16px; font-size:12px" class="meta"> <div style="display:flex; justify-content:space-between; margin-top:16px; font-size:12px" class="meta">
<span>~\${fit.req.toFixed(1)} GB RAM</span> <span>~${fit.req.toFixed(1)} GB RAM</span>
<span>\${m.quant}</span> <span>${m.quant}</span>
</div> </div>
</div> </div>
\`; `;
}).join(""); }).join("");
} }
@@ -259,12 +306,13 @@ window.openCuratedModal = (index) => {
$("#cb-modal").style.display = "flex"; $("#cb-modal").style.display = "flex";
$("#cb-m-title").textContent = m.name; $("#cb-m-title").textContent = m.name;
$("#cb-m-repo").textContent = m.repo; $("#cb-m-repo").textContent = m.repo;
$("#cb-m-files").innerHTML = \`<option value="\${esc(m.file)}">\${esc(m.file)}</option>\`; $("#cb-m-files").innerHTML = `<option value="${esc(m.file)}">${esc(m.file)}</option>`;
$("#cb-m-files").style.display = "block"; $("#cb-m-files").style.display = "block";
$("#cb-m-loading").style.display = "none"; $("#cb-m-loading").style.display = "none";
$("#cb-m-alias").value = m.alias; $("#cb-m-alias").value = m.alias;
$("#cb-m-ctx").value = m.ctx; $("#cb-m-ctx").value = m.ctx;
$("#cb-m-download").disabled = false; $("#cb-m-download").disabled = false;
updateLiveFit();
}; };
function onSystem(sys) { function onSystem(sys) {
+31 -10
View File
@@ -7,10 +7,13 @@ function render() {
if (!c) return; if (!c) return;
c.innerHTML = ` c.innerHTML = `
<div class="grid grid-2"> <div class="card-h"><h3>Guides & Integrationen</h3></div>
<div class="card"> <div class="card" style="padding:0; overflow:hidden;">
<div class="card-h"><h3>Cline / Cursor (Juni 2026)</h3></div>
<p>Nutze deine lokalen Modelle kostenlos in Cursor und Cline.</p> <details class="guide-acc">
<summary>Cline / Cursor (Juni 2026)</summary>
<div class="acc-body">
<p>Nutze deine lokalen Modelle kostenlos in der Cursor IDE und Cline.</p>
<label>Cursor: Settings -> Models -> Add Custom Model</label> <label>Cursor: Settings -> Models -> Add Custom Model</label>
<input class="tokin mono-sm" readonly value="coder"> <input class="tokin mono-sm" readonly value="coder">
<label>Base URL (OpenAI API)</label> <label>Base URL (OpenAI API)</label>
@@ -21,9 +24,11 @@ function render() {
In Cline: Wähle "OpenAI Compatible" als Provider. In Cline: Wähle "OpenAI Compatible" als Provider.
</p> </p>
</div> </div>
</details>
<div class="card"> <details class="guide-acc">
<div class="card-h"><h3>OpenWebUI (Pipes & Connections)</h3></div> <summary>OpenWebUI (Pipes & Connections)</summary>
<div class="acc-body">
<p>Verbinde llama-swap nativ in OpenWebUI.</p> <p>Verbinde llama-swap nativ in OpenWebUI.</p>
<label>Settings -> Admin Settings -> Connections</label> <label>Settings -> Admin Settings -> Connections</label>
<input class="tokin mono-sm" readonly value="${currentUrl}/v1"> <input class="tokin mono-sm" readonly value="${currentUrl}/v1">
@@ -33,9 +38,11 @@ function render() {
OpenWebUI erkennt nun vollautomatisch, wenn ein Modell ausgetauscht wird. OpenWebUI erkennt nun vollautomatisch, wenn ein Modell ausgetauscht wird.
</p> </p>
</div> </div>
</details>
<div class="card" style="grid-column: span 2;"> <details class="guide-acc">
<div class="card-h"><h3>Python / LangChain (OpenAI SDK)</h3></div> <summary>Python / LangChain (OpenAI SDK)</summary>
<div class="acc-body">
<p>Nutze das offizielle <code>openai</code> Package, um mit llama-swap zu sprechen.</p> <p>Nutze das offizielle <code>openai</code> Package, um mit llama-swap zu sprechen.</p>
<pre style="background: var(--bg); padding: 12px; border-radius: 8px; border: 1px solid var(--line); overflow-x: auto; color: var(--tx); font-family: monospace; font-size: 13px;"><code>from openai import OpenAI <pre style="background: var(--bg); padding: 12px; border-radius: 8px; border: 1px solid var(--line); overflow-x: auto; color: var(--tx); font-family: monospace; font-size: 13px;"><code>from openai import OpenAI
@@ -51,9 +58,11 @@ response = client.chat.completions.create(
print(response.choices[0].message.content)</code></pre> print(response.choices[0].message.content)</code></pre>
</div> </div>
</details>
<div class="card" style="grid-column: span 2;"> <details class="guide-acc">
<div class="card-h"><h3>n8n (AI Agent Nodes)</h3></div> <summary>n8n (AI Agent Nodes)</summary>
<div class="acc-body">
<p>Nutze den "OpenAI Chat Model" Node in n8n Advanced AI.</p> <p>Nutze den "OpenAI Chat Model" Node in n8n Advanced AI.</p>
<label>Credentials -> OpenAI API -> Custom URL</label> <label>Credentials -> OpenAI API -> Custom URL</label>
<input class="tokin mono-sm" readonly value="${currentUrl}/v1"> <input class="tokin mono-sm" readonly value="${currentUrl}/v1">
@@ -61,6 +70,18 @@ print(response.choices[0].message.content)</code></pre>
Einfach den Node verbinden, "coder" als Modell-ID (Expression) eintippen und loslegen. Einfach den Node verbinden, "coder" als Modell-ID (Expression) eintippen und loslegen.
</p> </p>
</div> </div>
</details>
<details class="guide-acc">
<summary>Begrifflichkeiten (Glossar)</summary>
<div class="acc-body">
<p><b>LLM-Engine (llama-swap):</b> Der Server im Hintergrund, der die Sprachmodelle (LLMs) lädt und OpenAI-kompatible Schnittstellen bereitstellt.</p>
<p><b>VRAM:</b> Der Grafikkarten-Speicher. KI-Modelle sind extrem groß und benötigen viel VRAM, um schnell zu laufen.</p>
<p><b>Quantisierung (z.B. Q4_K_M):</b> Ein Verfahren, das die Genauigkeit der Modellgewichte leicht reduziert (von 16-Bit auf 4-Bit), damit sie in den VRAM passen, ohne signifikant dümmer zu werden.</p>
<p><b>Context-Window:</b> Wie viel Text ("Tokens") sich das Modell gleichzeitig "merken" kann. Ein Buchstabe ist ~0.3 Tokens. Mehr Kontext braucht drastisch mehr VRAM.</p>
</div>
</details>
</div> </div>
`; `;
} }
+66 -13
View File
@@ -1,8 +1,10 @@
// models.js — "Modelle"-Ansicht: Download + Einpflegen, Schnelltest-Chat, Modell-Tabelle. // models.js — "Modelle"-Ansicht: Schnelltest-Chat, Modell-Tabelle & Konfiguration.
import { api } from "../core/api.js"; import { api } from "../core/api.js";
import { $, badge, esc, toast, icon } from "../core/ui.js"; import { $, badge, esc, toast, icon } from "../core/ui.js";
let ALL_MODELS = [];
function refreshSoon() { document.dispatchEvent(new Event("mc:refresh")); } function refreshSoon() { document.dispatchEvent(new Event("mc:refresh")); }
function mount() { function mount() {
@@ -13,43 +15,63 @@ function mount() {
<select id="chat-model"></select> <select id="chat-model"></select>
<label>Nachricht</label> <label>Nachricht</label>
<textarea id="chat-msg" placeholder="Schreib was, um ein Modell zu wecken…"></textarea> <textarea id="chat-msg" placeholder="Schreib was, um ein Modell zu wecken…"></textarea>
<button class="primary" id="chat-btn">Senden</button> <button class="primary" id="chat-btn" title="Sendet eine Chat-Anfrage an das Modell (weckt es auf, falls es schläft)">Senden</button>
<div id="chat-reply" class="reply" style="display:none"></div>`; <div id="chat-reply" class="reply" style="display:none"></div>`;
$("#m-table").innerHTML = ` $("#m-table").innerHTML = `
<div class="card-h"><h3>Modelle &amp; Ports</h3><span class="meta" id="m-count"></span></div> <div class="card-h"><h3>Modelle &amp; Ports</h3><span class="meta" id="m-count"></span></div>
<div class="hint" style="margin-bottom: 16px;"> <div class="hint" style="margin-bottom: 16px;">
💡 <b>Modelle werden automatisch geladen</b>, sobald eine Chat-Anfrage (z.B. über OpenWebUI oder Cursor) an sie gestellt wird. Du musst sie hier nicht manuell starten. 💡 <b>Modelle werden automatisch geladen</b>, sobald eine Chat-Anfrage an sie gestellt wird. Du musst sie nicht manuell starten.
</div> </div>
<table> <table>
<thead><tr><th>Modell</th><th>Fähigkeiten</th><th>Details</th><th>Status</th><th>Port</th><th style="text-align:right">Aktion</th></tr></thead> <thead><tr><th>Modell</th><th>Fähigkeiten</th><th>Details</th><th>Status</th><th>Port</th><th style="text-align:right">Aktionen</th></tr></thead>
<tbody id="models"></tbody> <tbody id="models"></tbody>
</table> </table>
<div id="models-empty" class="empty" style="display:none">Noch keine Modelle konfiguriert — zieh dir oben eins rein. 👇</div>`; <div id="models-empty" class="empty" style="display:none">Noch keine Modelle konfiguriert — zieh dir oben eins rein. 👇</div>
<!-- Config Modal -->
<div id="cfg-modal" style="display:none; position:fixed; top:0; left:0; right:0; bottom:0; background:rgba(0,0,0,0.8); z-index:100; align-items:center; justify-content:center;">
<div class="card" style="width:100%; max-width:400px; position:relative">
<button id="cfg-close" class="ghost" style="position:absolute; top:12px; right:12px;">Schließen</button>
<h2 style="margin-top:0">Modell konfigurieren</h2>
<p class="meta" id="cfg-model-name"></p>
<div style="margin-top:24px">
<label>Context-Size (Tokens)</label>
<input id="cfg-ctx" type="number" class="tokin" value="8192">
<div class="meta" style="font-size:12px; margin-top:4px;">Höhere Werte erlauben längere Dokumente, brauchen aber mehr VRAM.</div>
</div>
<button class="primary" id="cfg-save" style="width:100%; margin-top:24px; padding:12px">Speichern</button>
</div>
</div>
`;
$("#chat-btn").addEventListener("click", sendChat); $("#chat-btn").addEventListener("click", sendChat);
$("#cfg-close").addEventListener("click", () => $("#cfg-modal").style.display = "none");
$("#cfg-save").addEventListener("click", saveConfig);
} }
function onStatus(s) { function onStatus(s) {
const models = s?.models || []; ALL_MODELS = s?.models || [];
const tb = $("#models"); const tb = $("#models");
if (!tb) return; if (!tb) return;
tb.innerHTML = ""; tb.innerHTML = "";
$("#models-empty").style.display = models.length ? "none" : "block"; $("#models-empty").style.display = ALL_MODELS.length ? "none" : "block";
$("#m-count").textContent = models.length ? models.length + " konfiguriert" : ""; $("#m-count").textContent = ALL_MODELS.length ? ALL_MODELS.length + " konfiguriert" : "";
const sel = $("#chat-model"); const sel = $("#chat-model");
const cur = sel.value; const cur = sel.value;
sel.innerHTML = ""; sel.innerHTML = "";
for (const m of models) { for (const m of ALL_MODELS) {
const tr = document.createElement("tr"); const tr = document.createElement("tr");
let capsHtml = ""; let capsHtml = "";
if (m.meta && m.meta.caps) { if (m.meta && m.meta.caps) {
capsHtml = m.meta.caps.map(c => { capsHtml = m.meta.caps.map(c => {
if (c === "Text") return `<span class="meta" title="Text" style="display:inline-block;width:18px">${icon("search")}</span>`; if (c === "Text") return `<span class="meta" title="Text" style="display:inline-block;width:18px">${icon("compass")}</span>`;
if (c === "Code") return `<strong style="color:var(--blue);display:inline-block;width:18px" title="Code">${icon("code")}</strong>`; if (c === "Code") return `<strong style="color:var(--blue);display:inline-block;width:18px" title="Code">{ }</strong>`;
if (c === "Bild") return `<strong style="color:var(--purple);display:inline-block;width:18px" title="Vision">${icon("eye")}</strong>`; if (c === "Bild") return `<strong style="color:var(--purple);display:inline-block;width:18px" title="Vision">👁</strong>`;
return ""; return "";
}).join(" "); }).join(" ");
} }
@@ -71,14 +93,45 @@ function onStatus(s) {
<td>${detailsHtml}</td> <td>${detailsHtml}</td>
<td>${badge(m.state)}${perfHtml}</td> <td>${badge(m.state)}${perfHtml}</td>
<td class="port">${m.port ?? "auto"}</td> <td class="port">${m.port ?? "auto"}</td>
<td style="text-align:right"><button class="ghost" data-unload="${esc(m.name)}">Aus dem VRAM entfernen</button></td>`; <td style="text-align:right">
<button class="ghost" data-cfg="${esc(m.name)}" title="Context-Size anpassen">Konfigurieren</button>
<button class="ghost" data-unload="${esc(m.name)}" title="Beendet das Modell sofort und gibt den VRAM wieder frei">Aus VRAM entfernen</button>
</td>`;
tb.appendChild(tr); tb.appendChild(tr);
sel.insertAdjacentHTML("beforeend", `<option>${esc(m.name)}</option>`); sel.insertAdjacentHTML("beforeend", `<option>${esc(m.name)}</option>`);
} }
if (cur) sel.value = cur; if (cur) sel.value = cur;
tb.querySelectorAll("[data-unload]").forEach(b => tb.querySelectorAll("[data-unload]").forEach(b =>
b.addEventListener("click", () => unloadOne(b.getAttribute("data-unload"))) b.addEventListener("click", () => unloadOne(b.getAttribute("data-unload")))
); );
tb.querySelectorAll("[data-cfg]").forEach(b =>
b.addEventListener("click", () => openConfig(b.getAttribute("data-cfg")))
);
}
function openConfig(alias) {
const m = ALL_MODELS.find(x => x.name === alias);
if (!m) return;
$("#cfg-model-name").textContent = m.name;
$("#cfg-ctx").value = m.meta?.ctx || 8192;
$("#cfg-modal").style.display = "flex";
}
async function saveConfig() {
const alias = $("#cfg-model-name").textContent;
const ctx = parseInt($("#cfg-ctx").value) || 8192;
$("#cfg-save").disabled = true;
try {
await api("/api/update_model", { method: "POST", body: JSON.stringify({ alias, ctx }) });
toast("Gespeichert! Änderungen werden beim nächsten Modell-Start aktiv.");
$("#cfg-modal").style.display = "none";
refreshSoon();
} catch (e) {
toast(e.message, true);
}
$("#cfg-save").disabled = false;
} }
async function unloadOne(m) { async function unloadOne(m) {
+74 -25
View File
@@ -1,9 +1,10 @@
// overview.js — Dashboard-Kopf: Hero + kompakte Modell-Liste. // overview.js — Dashboard: Quick Actions, Modelle & Recent Jobs
import { $, esc } from "../core/ui.js"; import { $, esc, icon } from "../core/ui.js";
let S = null; // letzter Status let S = null; // letzter Status
let J = []; // letzte Job-Liste let J = []; // letzte Job-Liste
let SYS = null;
const RUNNING = new Set(["running", "ready", "loading", "starting"]); const RUNNING = new Set(["running", "ready", "loading", "starting"]);
@@ -12,34 +13,48 @@ function counts() {
return { return {
total: models.length, total: models.length,
running: models.filter(m => RUNNING.has(m.state)).length, running: models.filter(m => RUNNING.has(m.state)).length,
jobsRun: J.filter(j => j.state === "running" || j.state === "queued").length,
jobsErr: J.filter(j => j.state === "failed").length,
}; };
} }
function mini(label, val, tone = "") {
const v = tone ? `<b style="${tone === "bad" ? "color:var(--err)" : ""}">${val}</b>` : val;
return `<div class="mini"><div class="l">${label}</div><div class="v">${v}</div></div>`;
}
function renderHero() { function renderHero() {
const c = counts();
$("#hero").innerHTML = `<div class="hero"> $("#hero").innerHTML = `<div class="hero">
<div> <div>
<div class="eyebrow">Übersicht</div> <div class="eyebrow">Dashboard</div>
<h1>Mission Control</h1> <h1>Mission Control</h1>
<p>Steuerzentrale für deinen lokalen llama-swap-Stack — Modelle, Downloads, <p>Steuerzentrale für deinen lokalen llama-swap-Stack. Hier verwaltest du Modelle, Downloads und Server-Wartung.</p>
Wartung und Schnelltest an einem Ort.</p>
</div>
<div class="hero-stats">
${mini("Modelle", c.total)}
${mini("Aktiv", c.running, "on")}
${mini("Jobs", c.jobsRun)}
${mini("Fehler", c.jobsErr, c.jobsErr ? "bad" : "")}
</div> </div>
</div>`; </div>`;
} }
function renderQuickActions() {
// 3 Kacheln (Cookbook, Server-Status, Aktivität/Guides)
$("#ov-quick").innerHTML = `
<div class="card" style="cursor:pointer; display:flex; flex-direction:column; gap:12px; transition:border-color 0.2s" onclick="document.querySelector('.nav-item[data-view=\\'cookbook\\']').click()" onmouseover="this.style.borderColor='var(--act)'" onmouseout="this.style.borderColor='var(--line)'">
<div style="display:flex; justify-content:space-between; align-items:center;">
<h3 style="margin:0; font-size:16px;">Modell finden</h3>
<span style="color:var(--act)">${icon("book")}</span>
</div>
<p style="margin:0; font-size:13px; color:var(--mut);">Durchsuche HuggingFace nach neuen Modellen im Cookbook.</p>
</div>
<div class="card" style="cursor:pointer; display:flex; flex-direction:column; gap:12px; transition:border-color 0.2s" onclick="document.querySelector('.nav-item[data-view=\\'activity\\']').click()" onmouseover="this.style.borderColor='var(--act)'" onmouseout="this.style.borderColor='var(--line)'">
<div style="display:flex; justify-content:space-between; align-items:center;">
<h3 style="margin:0; font-size:16px;">Live Metriken</h3>
<span style="color:var(--act)">${icon("pulse")}</span>
</div>
<p style="margin:0; font-size:13px; color:var(--mut);">${SYS ? `System läuft (RAM: ${SYS.ram.percent.toFixed(0)}%, CPU: ${SYS.cpu.percent.toFixed(0)}%)` : 'Lade Metriken...'}</p>
</div>
<div class="card" style="cursor:pointer; display:flex; flex-direction:column; gap:12px; transition:border-color 0.2s" onclick="document.querySelector('.nav-item[data-view=\\'server\\']').click()" onmouseover="this.style.borderColor='var(--act)'" onmouseout="this.style.borderColor='var(--line)'">
<div style="display:flex; justify-content:space-between; align-items:center;">
<h3 style="margin:0; font-size:16px;">Wartung</h3>
<span style="color:var(--act)">${icon("server")}</span>
</div>
<p style="margin:0; font-size:13px; color:var(--mut);">Server neustarten, VRAM leeren oder Engine aktualisieren.</p>
</div>
`;
}
function modelRow(m) { function modelRow(m) {
const on = RUNNING.has(m.state); const on = RUNNING.has(m.state);
const dot = m.state === "loading" || m.state === "starting" ? "load" : on ? "on" : ""; const dot = m.state === "loading" || m.state === "starting" ? "load" : on ? "on" : "";
@@ -54,14 +69,15 @@ function modelRow(m) {
}).join(""); }).join("");
} }
const filename = m.meta?.filename ? `<div class="li-sub" style="font-family:var(--mono); color:var(--mut);">${esc(m.meta.filename)}</div>` : '';
return `<div class="li"> return `<div class="li">
<span class="li-dot ${dot}"></span> <span class="li-dot ${dot}"></span>
<div class="li-main"> <div class="li-main">
<div class="li-id" style="font-weight:500">${esc(m.name)}${caps}</div> <div class="li-id" style="font-weight:500">${esc(m.name)}${caps}</div>
<div class="li-sub">TTL ${m.ttl ?? "—"}${typeof m.ttl === "number" ? "s" : ""}</div> ${filename}
</div> </div>
<div class="li-right"> <div class="li-right">
<div class="li-meta">${m.port ?? "auto"}</div>
<div class="li-time">${state}</div> <div class="li-time">${state}</div>
</div> </div>
</div>`; </div>`;
@@ -77,10 +93,43 @@ function renderModels() {
<div class="e-s">Hol dir unter „Cookbook“ eins von HuggingFace.</div></div>`}`; <div class="e-s">Hol dir unter „Cookbook“ eins von HuggingFace.</div></div>`}`;
} }
function renderAll() { renderHero(); renderModels(); } function renderRecentJobs() {
const latest = J.slice(0, 4);
const statusBadge = (s) => {
if (s === "done") return '<span class="badge b-run" style="font-size:10px">fertig</span>';
if (s === "failed") return '<span class="badge b-err" style="font-size:10px">fehler</span>';
return '<span class="badge b-load" style="font-size:10px">lädt…</span>';
};
$("#ov-recent-jobs").innerHTML = `
<div class="card-h"><h3>Letzte Aktivitäten</h3><span class="meta" style="cursor:pointer" onclick="document.querySelector('.nav-item[data-view=\\'activity\\']').click()">Alle ansehen →</span></div>
${latest.length
? `<div class="list">
${latest.map(j => `
<div class="li" style="padding:10px 4px">
<div class="li-main">
<div class="li-id" style="font-size:12.5px">${esc(j.label)}</div>
</div>
<div class="li-right">${statusBadge(j.state)}</div>
</div>
`).join("")}
</div>`
: `<div class="empty-c"><div class="e-t">Keine Aktivitäten</div><div class="e-s">Alles läuft ruhig.</div></div>`
}
`;
}
function renderAll() {
renderHero();
renderQuickActions();
renderModels();
renderRecentJobs();
}
function mount() { renderAll(); } function mount() { renderAll(); }
function onStatus(s) { S = s; renderAll(); } function onStatus(s) { S = s; renderModels(); }
function onJobs(jobs) { J = jobs || []; renderHero(); } function onJobs(jobs) { J = jobs || []; renderRecentJobs(); }
function onSystem(sys) { SYS = sys; renderQuickActions(); }
export default { id: "overview", mount, onStatus, onJobs }; export default { id: "overview", mount, onStatus, onJobs, onSystem };
+6 -8
View File
@@ -7,13 +7,11 @@ function refreshSoon() { document.dispatchEvent(new Event("mc:refresh")); }
function mount() { function mount() {
$("#wartung").innerHTML = ` $("#wartung").innerHTML = `
<div class="card-h"><h3>Dienste & Applikation</h3></div> <div class="card-h"><h3>Dienste & Applikation</h3></div>
<div class="btn-row" style="margin-bottom:20px">
<button id="w-restart-swap">llama-swap neustarten</button>
<button id="w-restart-mc">Mission Control neustarten</button>
</div>
<div class="btn-row"> <div class="btn-row">
<button id="w-update">Mission Control System-Update</button> <button id="w-restart-swap" title="Startet die LLM-Engine neu. Dauer ca. 5 Sekunden.">llama-swap neustarten</button>
<button id="w-unload" class="ghost">Aus dem VRAM entfernen</button> <button id="w-restart-mc" title="Startet Mission Control (das Dashboard) neu.">Mission Control neustarten</button>
<button id="w-update" title="Zieht den neuesten Code via Git und aktualisiert das Dashboard.">Mission Control System-Update</button>
<button id="w-unload" class="ghost" title="Wirft alle derzeit in VRAM geladenen Modelle sofort raus.">Aus dem VRAM entfernen</button>
</div> </div>
<div class="hint" style="margin-top:12px; margin-bottom:32px"> <div class="hint" style="margin-top:12px; margin-bottom:32px">
Dienste starten via passwortlosem Sudo neu. Dienste starten via passwortlosem Sudo neu.
@@ -21,8 +19,8 @@ function mount() {
<div class="card-h"><h3>Betriebssystem (Bosgame)</h3></div> <div class="card-h"><h3>Betriebssystem (Bosgame)</h3></div>
<div class="btn-row"> <div class="btn-row">
<button id="w-os-update">OS-Updates installieren (apt update)</button> <button id="w-os-update" title="Führt apt update & upgrade aus, um das Betriebssystem (Bosgame) zu aktualisieren.">OS-Updates installieren (apt update)</button>
<button id="w-reboot" class="danger">Server Reboot</button> <button id="w-reboot" class="danger" title="Startet den gesamten Server physikalisch neu. Alles ist kurz offline.">Server Reboot</button>
</div> </div>
<div class="hint" style="margin-top:12px; margin-bottom:32px"> <div class="hint" style="margin-top:12px; margin-bottom:32px">
Für tiefe Eingriffe fragt das Dashboard einmalig das sudo-Passwort ab. Für tiefe Eingriffe fragt das Dashboard einmalig das sudo-Passwort ab.