Free, open-source AI video generator for Mac. Run LTX-2.5 and MiniMax H3 (Hailuo) locally on Apple Silicon: text-to-video and image-to-video with sound, lip-sync, character LoRA training, songs from lyrics (YuE2) and image editing (Qwen-Image-Edit). One-click install via Pinokio. Offline: no cloud, no API key, no NVIDIA GPU.
Python ai-filmmakingai-videoai-video-generatorapple-siliconhailuoimage-to-videolip-synclocal-ailora-trainingltx-2ltx-2-5ltx-videomacosminimax-h3mlxmusic-generationoffline-aipinokiotext-to-videovideo-generation
feat(panel/train): promote Train to workflow tier (top tab strip)
mrbizarro682b88ee8327
diff --git a/mlx_ltx_panel.py b/mlx_ltx_panel.pyindex 7f64e9e..8962068 100644--- a/mlx_ltx_panel.py+++ b/mlx_ltx_panel.py@@ -9171,6 +9171,18 @@ HTML = r"""<!doctype html>body[data-workflow="agent"] #genForm,body[data-workflow="agent"] #studioSection { display: none !important; }+ /* Train Character is its own workflow tier alongside Manual + Agent.+ Hide the manual-only chrome (Mode pills, models card, genForm, studio)+ so only #trainSection is visible inside form-pane. The stage-pane+ stays visible — training surfaces progress through the existing Now /+ Queue / Recent / Logs panes. */+ body[data-workflow="train"] #modelsInline,+ body[data-workflow="train"] #modeGroup,+ body[data-workflow="train"] aside.form-pane > h2,+ body[data-workflow="train"] #genForm,+ body[data-workflow="train"] #studioSection { display: none !important; }+ body[data-workflow="train"] #trainSection { display: block; }+/* ===== FORM ===== */h2 {font-size: 10px; margin: 14px 0 8px; color: var(--muted);@@ -17030,6 +17042,7 @@ HTML = r"""<!doctype html><nav class="workflow-tabs" id="workflowTabs"><button data-workflow="manual" class="active">Manual</button><button data-workflow="agent">Agentic Flows<span class="new-badge">NEW</span></button>+ <button data-workflow="train">Train<span class="new-badge">NEW</span></button></nav><!-- ============== AGENT SESSIONS SIDEBAR ============== -->@@ -17338,7 +17351,9 @@ HTML = r"""<!doctype html><button type="button" class="mode-chip pill-btn" data-mode="keyframe">FFLF<span class="mc-sub sub">first + last</span></button><button type="button" class="mode-chip pill-btn" data-mode="extend">Extend<span class="mc-sub sub">continue a clip</span></button><button type="button" class="mode-chip pill-btn" data-mode="image" title="Open Image Studio — generate stills (Qwen-Image-Edit-2511 multi-ref + others)">Studio<span class="mc-sub sub">stills + library</span></button>- <button type="button" class="mode-chip pill-btn" data-mode="train" title="Train a character LoRA from 15-50 stills — runs in the sibling lora-lab pipeline">Train<span class="mc-sub sub">character LoRA</span></button>+ <!-- "Train" used to live here as a mode chip; it's been promoted to a+ workflow tier (top tab strip, alongside Manual / Agentic Flows)+ because training is a different category than rendering. --></div><form id="genForm">@@ -24103,6 +24118,7 @@ function workflowSwitch(name) {const manual = document.getElementById('genForm');const agent = document.getElementById('agentPane');const studio = document.getElementById('studioSection');+ const train = document.getElementById('trainSection');// Set body data attribute so CSS can switch the layout (wider form-pane,// show agent-stage-pane on the right).document.body.setAttribute('data-workflow', name);@@ -24113,6 +24129,7 @@ function workflowSwitch(name) {// browser-driven audit: switching tabs while Studio was active left// the chat inaccessible.if (studio) studio.classList.remove('show');+ if (train) train.classList.remove('show');if (name === 'agent') {if (manual) manual.style.display = 'none';if (agent) agent.hidden = false;@@ -24129,6 +24146,15 @@ function workflowSwitch(name) {const ta = document.getElementById('agentInput');if (ta) { agentAutoResize(ta); ta.focus(); }}, 50);+ } else if (name === 'train') {+ // Train Character is its own workflow tier (alongside Manual and+ // Agentic Flows). Hide the manual render form + agent chat, show+ // the train section, run its init.+ if (manual) manual.style.display = 'none';+ if (agent) agent.hidden = true;+ agentStageStop();+ if (train) train.classList.add('show');+ if (typeof trainInit === 'function') trainInit();} else {if (manual) manual.style.display = '';if (agent) agent.hidden = true;