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
fix(ui): polish pass — bubbles, tokens, SVG icons, focus ring
mrbizarro263bc6c3a53d
diff --git a/mlx_ltx_panel.py b/mlx_ltx_panel.pyindex 83955d5..2a110c5 100644--- a/mlx_ltx_panel.py+++ b/mlx_ltx_panel.py@@ -5801,6 +5801,17 @@ HTML = r"""<!doctype html>--accent: #2f81f7; --accent-bright: #58a6ff; --accent-dim: rgba(47,129,247,0.18);--success: #3fb950; --warning: #d29922; --danger: #f85149;--radius: 10px;+ /* Polish tokens — adopted across the app for one consistent+ elevation + radius + motion language. Pre-token, every component+ hand-rolled its own shadow / radius / transition so the chrome+ drifted into "Pythonic" territory. */+ --shadow-1: 0 1px 2px rgba(0,0,0,0.30), 0 1px 0 rgba(255,255,255,0.02) inset;+ --shadow-2: 0 8px 24px rgba(0,0,0,0.40);+ --ring: 0 0 0 3px var(--accent-dim);+ --r-xs: 6px; --r-sm: 8px; --r-md: 10px; --r-lg: 14px; --r-pill: 999px;+ --t-fast: 120ms cubic-bezier(.2,.8,.2,1);+ --t-base: 180ms cubic-bezier(.2,.8,.2,1);+ --t-slow: 240ms cubic-bezier(.2,.8,.2,1);}* { box-sizing: border-box; }html, body { margin: 0; height: 100%; }@@ -6169,7 +6180,12 @@ HTML = r"""<!doctype html>width: 100%; padding: 8px 11px; font: inherit; color: inherit;background: var(--panel-2); border: 1px solid var(--border); border-radius: 6px;}- input:focus, textarea:focus, select:focus { outline: none; border-color: var(--accent); background: var(--bg-2); }+ input:focus, textarea:focus, select:focus {+ outline: none;+ border-color: var(--accent);+ background: var(--bg-2);+ box-shadow: var(--ring);+ }textarea { min-height: 84px; resize: vertical; font-family: inherit; }textarea.avoid-textarea { min-height: 54px; }@@ -7316,8 +7332,9 @@ HTML = r"""<!doctype html>.workflow-tabs .new-badge {display: inline-block; margin-left: 6px;font-size: 9px; font-weight: 700; letter-spacing: 0.5px;- padding: 2px 6px; border-radius: 999px;- background: var(--accent); color: white;+ padding: 2px 6px; border-radius: var(--r-pill);+ background: var(--accent-dim); color: var(--accent-bright);+ border: 1px solid var(--accent);vertical-align: middle;}@@ -7388,8 +7405,10 @@ HTML = r"""<!doctype html>background: var(--accent-dim, rgba(47,129,247,0.12));}.agent-mode-icon {- font-size: 13px; line-height: 1;- filter: saturate(0.85);+ display: inline-flex;+ align-items: center; justify-content: center;+ line-height: 0;+ flex-shrink: 0;}/* Stop-engine quick action — same height as the pills, danger coloron hover so it reads as "this kills something". Only shown when@@ -7913,6 +7932,21 @@ HTML = r"""<!doctype html>color: var(--text);word-wrap: break-word; overflow-wrap: break-word;}+ /* Distinguish user vs assistant bubbles. Pre-this commit both sides+ rendered as naked text on the void, so the chat read like one+ wall of dev-tool output. User now gets a tinted bubble with the+ chat-app tail-truncated radius (Claude.ai style); assistant+ stays naked text on the void so longer planning replies don't+ feel boxed-in. */+ .agent-msg-row.user .agent-msg-content {+ background: var(--panel-2);+ border: 1px solid var(--border);+ border-radius: var(--r-lg) var(--r-lg) var(--r-xs) var(--r-lg);+ padding: 9px 14px;+ }+ .agent-msg-row.user .agent-msg-attachments {+ margin-top: 2px;+ }/* ---- Markdown rendering inside .agent-md ---- */.agent-md > *:first-child { margin-top: 0; }@@ -9675,7 +9709,7 @@ HTML = r"""<!doctype html><button type="button" class="agent-mode-pill" id="agentModePill"onclick="agentToggleMode()"title="Plan & sleep: auto-stop the chat model after the agent finishes (frees RAM for renders). Click to switch to Interactive mode.">- <span class="agent-mode-icon" id="agentModeIcon">🌙</span>+ <span class="agent-mode-icon" id="agentModeIcon"></span><span id="agentModeLabel">Plan & sleep</span></button><!-- Quick "stop engine" — only shown when the local engine is@@ -14054,14 +14088,24 @@ function agentRenderModePill() {if (!pill) return;const mode = ((window.AGENT.config && window.AGENT.config.engine&& window.AGENT.config.engine.mode) || 'plan_sleep');+ // Monochrome stroked SVGs — pro-app feel. Color emoji glyphs read+ // as Pythonic in chrome that's otherwise stroke-icon based.+ const ICON_INTERACTIVE =+ '<svg viewBox="0 0 24 24" width="13" height="13" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round">' ++ '<path d="M21 15a2 2 0 0 1-2 2H7l-4 4V5a2 2 0 0 1 2-2h14a2 2 0 0 1 2 2z"/>' ++ '</svg>';+ const ICON_PLAN_SLEEP =+ '<svg viewBox="0 0 24 24" width="13" height="13" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round">' ++ '<path d="M21 12.79A9 9 0 1 1 11.21 3 7 7 0 0 0 21 12.79z"/>' ++ '</svg>';if (mode === 'interactive') {pill.classList.add('is-interactive');- if (icon) icon.textContent = '💬';+ if (icon) icon.innerHTML = ICON_INTERACTIVE;if (label) label.textContent = 'Interactive';pill.title = 'Interactive: chat model stays resident across finishes (uses ~22 GB for Qwen 35B).\nClick to switch to Plan & sleep mode.';} else {pill.classList.remove('is-interactive');- if (icon) icon.textContent = '🌙';+ if (icon) icon.innerHTML = ICON_PLAN_SLEEP;if (label) label.textContent = 'Plan & sleep';pill.title = 'Plan & sleep: auto-stop the chat model after the agent finishes (frees RAM for renders).\nClick to switch to Interactive mode.';}@@ -14776,7 +14820,7 @@ function renderMessage(m) {if (m.kind === 'system_note') return renderSystemNote(m.content || '');const row = document.createElement('div');- row.className = 'agent-msg-row';+ row.className = 'agent-msg-row ' + (m.kind === 'user' ? 'user' : 'assistant');const av = document.createElement('div');av.className = `agent-avatar ${m.kind === 'user' ? 'user' : 'claude'}`;