diff --git a/.dockerignore b/.dockerignore
new file mode 100644
index 00000000..f3166178
--- /dev/null
+++ b/.dockerignore
@@ -0,0 +1,9 @@
+node_modules
+python-embed
+dist
+dist-electron
+release
+backend/.venv
+backend/outputs
+backend/models
+.git
diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
index df5dddac..ea58a42a 100644
--- a/.github/workflows/ci.yml
+++ b/.github/workflows/ci.yml
@@ -59,6 +59,8 @@ jobs:
             runner: macos-latest
           - os: windows
             runner: windows-latest
+          - os: linux
+            runner: ubuntu-latest
     defaults:
       run:
         working-directory: backend
diff --git a/AGENTS.md b/AGENTS.md
index 5a634664..92828aea 100644
--- a/AGENTS.md
+++ b/AGENTS.md
@@ -21,8 +21,8 @@ LTX Desktop is an Electron app for AI video generation using LTX models. Three-l
 | `pnpm typecheck:py` | Python pyright only |
 | `pnpm backend:test` | Run Python pytest tests |
 | `pnpm build:frontend` | Vite frontend build only |
-| `pnpm build:mac` / `pnpm build:win` | Full platform builds |
-| `pnpm setup:dev:mac` / `pnpm setup:dev:win` | One-time dev environment setup |
+| `pnpm build` | Full platform build (auto-detects platform) |
+| `pnpm setup:dev` | One-time dev environment setup (auto-detects platform) |
 
 Run a single backend test file via pnpm: `pnpm backend:test -- tests/test_ic_lora.py`
 
diff --git a/README.md b/README.md
index acc7f4b2..a5453180 100644
--- a/README.md
+++ b/README.md
@@ -1,6 +1,6 @@
 # LTX Desktop
 
-LTX Desktop is an open-source desktop app for generating videos with LTX models — locally on supported Windows NVIDIA GPUs, with an API mode for unsupported hardware and macOS.
+LTX Desktop is an open-source desktop app for generating videos with LTX models — locally on supported Windows/Linux NVIDIA GPUs, with an API mode for unsupported hardware and macOS.
 
 > **Status: Beta.** Expect breaking changes.
 > Frontend architecture is under active refactor; large UI PRs may be declined for now (see [`CONTRIBUTING.md`](docs/CONTRIBUTING.md)).
@@ -32,8 +32,9 @@ LTX Desktop is an open-source desktop app for generating videos with LTX models
 | --- | --- | --- |
 | Windows + CUDA GPU with **≥32GB VRAM** | Local generation | Downloads model weights locally |
 | Windows (no CUDA, <32GB VRAM, or unknown VRAM) | API-only | **LTX API key required** |
+| Linux + CUDA GPU with **≥32GB VRAM** | Local generation | Downloads model weights locally |
+| Linux (no CUDA, <32GB VRAM, or unknown VRAM) | API-only | **LTX API key required** |
 | macOS (Apple Silicon builds) | API-only | **LTX API key required** |
-| Linux | Not officially supported | No official builds |
 
 In API-only mode, available resolutions/durations may be limited to what the API supports.
 
@@ -46,6 +47,14 @@ In API-only mode, available resolutions/durations may be limited to what the API
 - 16GB+ RAM (32GB recommended)
 - **160GB+ free disk space** (for model weights, Python environment, and outputs)
 
+### Linux (local generation)
+
+- Ubuntu 22.04+ or similar distro (x64 or arm64)
+- NVIDIA GPU with CUDA support and **≥32GB VRAM** (more is better)
+- NVIDIA driver installed (PyTorch bundles the CUDA runtime)
+- 16GB+ RAM (32GB recommended)
+- Plenty of free disk space for model weights and outputs
+
 ### macOS (API-only)
 
 - Apple Silicon (arm64)
@@ -64,6 +73,7 @@ LTX Desktop stores app data (settings, models, logs) in:
 
 - **Windows:** `%LOCALAPPDATA%\LTXDesktop\`
 - **macOS:** `~/Library/Application Support/LTXDesktop/`
+- **Linux:** `$XDG_DATA_HOME/LTXDesktop/` (default: `~/.local/share/LTXDesktop/`)
 
 Model weights are downloaded into the `models/` subfolder (this can be large and may take time).
 
@@ -84,7 +94,7 @@ The LTX API is used for:
 - API-based video generations (required on macOS and on unsupported Windows hardware) — paid
 - Retake — paid
 
-An LTX API key is required in API-only mode, but optional on Windows local mode if you enable the Local Text Encoder.
+An LTX API key is required in API-only mode, but optional on Windows/Linux local mode if you enable the Local Text Encoder.
 
 Generate a FREE API key at the [LTX Console](https://console.ltx.video/). Text encoding is free; video generation API usage is paid. [Read more](https://ltx.io/model/model-blog/ltx-2-better-control-for-real-workflows).
 
@@ -137,11 +147,7 @@ Prereqs:
 Setup:
 
 ```bash
-# macOS
-pnpm setup:dev:mac
-
-# Windows
-pnpm setup:dev:win
+pnpm setup:dev
 ```
 
 Run:
diff --git a/backend/pyproject.toml b/backend/pyproject.toml
index 797e3bf8..a546d244 100644
--- a/backend/pyproject.toml
+++ b/backend/pyproject.toml
@@ -27,6 +27,7 @@ dependencies = [
     "uvicorn[standard]>=0.30.0",
     "python-multipart>=0.0.9",
     "triton-windows; sys_platform == 'win32'",
+    "triton; sys_platform == 'linux'",
 ]
 
 [[tool.uv.index]]
diff --git a/backend/runtime_config/runtime_policy.py b/backend/runtime_config/runtime_policy.py
index cf64b66b..7b84dc89 100644
--- a/backend/runtime_config/runtime_policy.py
+++ b/backend/runtime_config/runtime_policy.py
@@ -8,7 +8,7 @@ def decide_force_api_generations(system: str, cuda_available: bool, vram_gb: int
     if system == "Darwin":
         return True
 
-    if system == "Windows":
+    if system in ("Windows", "Linux"):
         if not cuda_available:
             return True
         if vram_gb is None:
diff --git a/backend/tests/test_runtime_policy_decision.py b/backend/tests/test_runtime_policy_decision.py
index 0999f709..f4f8ba00 100644
--- a/backend/tests/test_runtime_policy_decision.py
+++ b/backend/tests/test_runtime_policy_decision.py
@@ -26,5 +26,21 @@ def test_windows_with_required_vram_allows_local_mode() -> None:
     assert decide_force_api_generations(system="Windows", cuda_available=True, vram_gb=31) is False
 
 
+def test_linux_without_cuda_forces_api() -> None:
+    assert decide_force_api_generations(system="Linux", cuda_available=False, vram_gb=24) is True
+
+
+def test_linux_with_low_vram_forces_api() -> None:
+    assert decide_force_api_generations(system="Linux", cuda_available=True, vram_gb=30) is True
+
+
+def test_linux_with_unknown_vram_forces_api() -> None:
+    assert decide_force_api_generations(system="Linux", cuda_available=True, vram_gb=None) is True
+
+
+def test_linux_with_required_vram_allows_local_mode() -> None:
+    assert decide_force_api_generations(system="Linux", cuda_available=True, vram_gb=31) is False
+
+
 def test_other_systems_fail_closed() -> None:
-    assert decide_force_api_generations(system="Linux", cuda_available=True, vram_gb=48) is True
+    assert decide_force_api_generations(system="FreeBSD", cuda_available=True, vram_gb=48) is True
diff --git a/backend/uv.lock b/backend/uv.lock
index 07a19c05..c49d255e 100644
--- a/backend/uv.lock
+++ b/backend/uv.lock
@@ -531,6 +531,7 @@ dependencies = [
     { name = "torch", version = "2.10.0+cu128", source = { registry = "https://download.pytorch.org/whl/cu128" }, marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
     { name = "tqdm" },
     { name = "transformers" },
+    { name = "triton", marker = "sys_platform == 'linux'" },
     { name = "triton-windows", marker = "sys_platform == 'win32'" },
     { name = "uvicorn", extra = ["standard"] },
 ]
@@ -574,6 +575,7 @@ requires-dist = [
     { name = "torch", marker = "sys_platform == 'linux' or sys_platform == 'win32'", specifier = ">=2.3.0", index = "https://download.pytorch.org/whl/cu128" },
     { name = "tqdm", specifier = ">=4.66.0" },
     { name = "transformers", specifier = ">=4.52,<5" },
+    { name = "triton", marker = "sys_platform == 'linux'" },
     { name = "triton-windows", marker = "sys_platform == 'win32'" },
     { name = "uvicorn", extras = ["standard"], specifier = ">=0.30.0" },
 ]
diff --git a/docs/CONTRIBUTING.md b/docs/CONTRIBUTING.md
index a642e8bb..b0f9dedf 100644
--- a/docs/CONTRIBUTING.md
+++ b/docs/CONTRIBUTING.md
@@ -14,11 +14,7 @@ Prereqs:
 Setup:
 
 ```bash
-# macOS
-pnpm setup:dev:mac
-
-# Windows
-pnpm setup:dev:win
+pnpm setup:dev
 ```
 
 Run:
diff --git a/docs/INSTALLER.md b/docs/INSTALLER.md
index fce6f70f..f501b8df 100644
--- a/docs/INSTALLER.md
+++ b/docs/INSTALLER.md
@@ -10,7 +10,7 @@ This guide explains how to build a distributable installer for **LTX Desktop**.
 The installer includes:
 - **Electron app** (React frontend + Electron shell)
 - **Embedded Python** (version from [`backend/.python-version`](../backend/.python-version)) with all dependencies pre-installed:
-  - PyTorch (CUDA on Windows, MPS on macOS)
+  - PyTorch (CUDA on Windows/Linux, MPS on macOS)
   - FastAPI, Diffusers, Transformers
   - LTX-2 inference packages
   - All other required libraries
@@ -18,6 +18,7 @@ The installer includes:
 
 **NOT bundled** (downloaded at runtime):
 - Model weights (downloaded on first run; can be large) from Hugging Face
+- On **Linux** and **Windows**: the Python environment itself is downloaded on first launch (keeps installer small)
 
 The embedded Python is **fully isolated** from the target system's Python — it lives inside `{install_dir}/resources/python/` and never modifies system settings.
 
@@ -35,67 +36,43 @@ Before building, ensure you have:
 
 - **Windows**: PowerShell 5.1+ (comes with Windows 10/11)
 - **macOS**: Xcode Command Line Tools (`xcode-select --install`)
+- **Linux**: `build-essential` (or equivalent) for native extensions
 
 ## Quick Build
 
-### macOS
 ```bash
-pnpm build:mac
-```
-
-### Windows
-```powershell
-pnpm build:win
+pnpm build
 ```
 
-This will:
+This auto-detects your platform and will:
 1. Download a standalone Python distribution (version from [`backend/.python-version`](../backend/.python-version))
-2. Install all Python dependencies (~10GB on Windows with CUDA, ~2-3GB on macOS with MPS)
+2. Install all Python dependencies (~10GB on Windows/Linux with CUDA, ~2-3GB on macOS with MPS)
 3. Build the frontend
 4. Package everything with electron-builder
-5. Create a DMG (macOS) or NSIS installer (Windows) in the `release/` folder
+5. Create a DMG (macOS), AppImage + deb (Linux), or NSIS installer (Windows) in the `release/` folder
 
 ## Build Options
 
-### macOS
-
 ```bash
 # Full build
-pnpm build:mac
+pnpm build
 
 # Skip Python setup (if already prepared)
-pnpm build:mac:skip-python
+pnpm build:skip-python
 
 # Fast rebuild (unpacked, skip Python + pnpm install)
-pnpm build:fast:mac
+pnpm build:fast
 
 # Just prepare Python environment
-pnpm prepare:python:mac
+pnpm prepare:python
 ```
 
-### Windows
-
-```powershell
-# Full build
-pnpm build:win
-
-# Skip Python setup (if already prepared)
-pnpm build:win:skip-python
-
-# Just prepare Python environment
-pnpm prepare:python:win
-
-# Fast rebuild (unpacked, skip Python + pnpm install)
-pnpm build:fast:win
-
-# Clean build
-powershell -File scripts/local-build.ps1 -Clean
-```
+All commands auto-detect the current platform (macOS, Linux, or Windows).
 
 ### Build Script Options
 
-The `local-build.sh` script accepts:
-- `--platform mac|win` — Target platform (auto-detected if omitted)
+The underlying `local-build.sh` / `local-build.ps1` scripts also accept:
+- `--platform mac|linux|win` — Target platform (auto-detected if omitted)
 - `--skip-python` — Use existing `python-embed/` directory
 - `--clean` — Remove build artifacts before starting
 - `--unpack` — Build unpacked app only (faster, no installer/DMG)
@@ -108,6 +85,15 @@ release/
   └── LTX Desktop-<version>-arm64.dmg
 ```
 
+### Linux
+```
+release/
+  ├── LTX Desktop-x86_64.AppImage
+  ├── LTX Desktop-amd64.deb
+  ├── LTX Desktop-arm64.AppImage
+  └── LTX Desktop-arm64.deb
+```
+
 ### Windows
 ```
 release/
@@ -118,7 +104,7 @@ release/
 
 Place icon files in `resources/` before building:
 - `icon.ico` — Windows (multi-size ICO: 256x256, 128x128, 64x64, 48x48, 32x32, 16x16)
-- `icon.png` — macOS (1024x1024 recommended)
+- `icon.png` — macOS and Linux (1024x1024 recommended)
 
 ## Troubleshooting
 
@@ -137,6 +123,7 @@ xattr -dr com.apple.quarantine /Applications/LTX\ Desktop.app
 ### Installer is too large
 Expected installer sizes (does not include model weights):
 - **Windows**: ~10GB (PyTorch CUDA ~2.5GB + ML libraries ~5GB + Python ~200MB + Electron ~100MB)
+- **Linux**: ~10GB (similar to Windows; PyTorch CUDA variant)
 - **macOS**: ~2-3GB (PyTorch MPS is much smaller than CUDA variant)
 
 ### Runtime / first-run issues
@@ -156,10 +143,28 @@ pnpm install
 pnpm build:frontend
 
 # 4. Build DMG
-npx electron-builder --mac
+pnpm exec electron-builder --mac
+
+# Or build unpacked app (faster, for testing)
+pnpm exec electron-builder --mac --dir
+```
+
+### Linux
+```bash
+# 1. Prepare Python environment
+bash scripts/prepare-python.sh
+
+# 2. Install dependencies
+pnpm install
+
+# 3. Build frontend
+pnpm build:frontend
+
+# 4. Build AppImage + deb
+pnpm exec electron-builder --linux
 
 # Or build unpacked app (faster, for testing)
-npx electron-builder --mac --dir
+pnpm exec electron-builder --linux --dir
 ```
 
 ### Windows
@@ -174,5 +179,5 @@ pnpm install
 pnpm build:frontend
 
 # 4. Build installer
-npx electron-builder --win
+pnpm exec electron-builder --win
 ```
diff --git a/docs/TELEMETRY.md b/docs/TELEMETRY.md
index b1e8a3eb..02324a0a 100644
--- a/docs/TELEMETRY.md
+++ b/docs/TELEMETRY.md
@@ -17,6 +17,7 @@ App data folder locations:
 
 - **Windows:** `%LOCALAPPDATA%\LTXDesktop\`
 - **macOS:** `~/Library/Application Support/LTXDesktop/`
+- **Linux:** `$XDG_DATA_HOME/LTXDesktop/` (default: `~/.local/share/LTXDesktop/`)
 
 Your preference is respected immediately — no restart required.
 
diff --git a/electron-builder.yml b/electron-builder.yml
index 94753cc6..6fb6ff23 100644
--- a/electron-builder.yml
+++ b/electron-builder.yml
@@ -92,6 +92,26 @@ dmg:
       type: link
       path: /Applications
 
+linux:
+  target:
+    - AppImage
+    - deb
+  icon: resources/icons
+  category: Video
+  artifactName: ${productName}-${arch}.${ext}
+
+deb:
+  depends:
+    - libgtk-3-0
+    - libnotify4
+    - libnss3
+    - libxss1
+    - libxtst6
+    - xdg-utils
+    - libatspi2.0-0
+    - libuuid1
+    - libsecret-1-0
+
 publish:
   provider: github
   owner: Lightricks
diff --git a/electron/app-paths.ts b/electron/app-paths.ts
index ddf10e6f..40a36115 100644
--- a/electron/app-paths.ts
+++ b/electron/app-paths.ts
@@ -18,7 +18,8 @@ function resolveUserDataPath(): string {
       APP_FOLDER_NAME,
     )
   }
-  return path.join(os.homedir(), `.${APP_FOLDER_NAME}`)
+  const xdgData = process.env.XDG_DATA_HOME || path.join(os.homedir(), '.local', 'share')
+  return path.join(xdgData, APP_FOLDER_NAME)
 }
 
 app.setPath('userData', resolveUserDataPath())
diff --git a/electron/export/ffmpeg-utils.ts b/electron/export/ffmpeg-utils.ts
index c8f4f631..cf511276 100644
--- a/electron/export/ffmpeg-utils.ts
+++ b/electron/export/ffmpeg-utils.ts
@@ -3,6 +3,7 @@ import path from 'path'
 import fs from 'fs'
 import { isDev, getCurrentDir } from '../config'
 import { logger } from '../logger'
+import { getPythonDir } from '../python-setup'
 
 let activeExportProcess: ChildProcess | null = null
 
@@ -13,12 +14,12 @@ export function findFfmpegPath(): string | null {
     const imageioRelPath = path.join('Lib', 'site-packages', 'imageio_ffmpeg', 'binaries')
     binDir = isDev
       ? path.join(getCurrentDir(), 'backend', '.venv', imageioRelPath)
-      : path.join(process.resourcesPath, 'python', imageioRelPath)
+      : path.join(getPythonDir(), imageioRelPath)
   } else {
     // macOS/Linux: find lib/python3.X/site-packages dynamically
     const venvBase = isDev
       ? path.join(getCurrentDir(), 'backend', '.venv')
-      : path.join(process.resourcesPath, 'python')
+      : getPythonDir()
     const libDir = path.join(venvBase, 'lib')
     if (fs.existsSync(libDir)) {
       const pythonDir = fs.readdirSync(libDir).find(e => e.startsWith('python3'))
diff --git a/electron/python-setup.ts b/electron/python-setup.ts
index c429b4d3..19f59092 100644
--- a/electron/python-setup.ts
+++ b/electron/python-setup.ts
@@ -66,7 +66,7 @@ function getInstalledHashPath(): string {
 
 /** Directory where python-embed lives at runtime. */
 export function getPythonDir(): string {
-  if (process.platform === 'win32') {
+  if (process.platform === 'win32' || process.platform === 'linux') {
     if (isDev) {
       return path.join(process.cwd(), 'python-embed')
     }
@@ -81,7 +81,7 @@ export function getPythonDir(): string {
  * Also promotes a staged python-next/ directory if it matches the expected hash.
  */
 export function isPythonReady(): { ready: boolean } {
-  if (process.platform !== 'win32') {
+  if (process.platform === 'darwin') {
     return { ready: true }
   }
 
@@ -112,7 +112,7 @@ export function isPythonReady(): { ready: boolean } {
   const installedHash = readHash(getInstalledHashPath())
 
   if (!bundledHash) {
-    const pythonExe = path.join(getPythonDir(), 'python.exe')
+    const pythonExe = path.join(getPythonDir(), process.platform === 'win32' ? 'python.exe' : 'bin/python3')
     return { ready: fs.existsSync(pythonExe) }
   }
 
@@ -128,7 +128,7 @@ export async function preDownloadPythonForUpdate(
   newVersion: string,
   onProgress?: (progress: PythonSetupProgress) => void
 ): Promise<boolean> {
-  if (process.platform !== 'win32') {
+  if (process.platform === 'darwin') {
     return false
   }
 
@@ -190,7 +190,8 @@ export async function preDownloadPythonForUpdate(
     try {
       await acquireArchive(baseUrl, archivePath, cleanupFiles, progressCb)
     } catch (primaryErr) {
-      const fallbackUrl = newHash ? `${FALLBACK_CDN_BASE}/python-embed-win32/${newHash}/python-embed-win32.tar.gz` : null
+      const prefix = getPythonArchivePrefix()
+      const fallbackUrl = newHash ? `${FALLBACK_CDN_BASE}/${prefix}/${newHash}/${prefix}.tar.gz` : null
       if (!fallbackUrl || isLocalPath(baseUrl)) {
         throw primaryErr
       }
@@ -242,6 +243,16 @@ function readHash(filePath: string): string | null {
 
 const FALLBACK_CDN_BASE = 'https://storage.googleapis.com/ltx-desktop-artifacts'
 
+function getPythonArchivePrefix(): string {
+  if (process.platform === 'win32') return 'python-embed-win32'
+  if (process.platform === 'linux') {
+    if (process.arch === 'x64') return 'python-embed-linux-x64'
+    if (process.arch === 'arm64') return 'python-embed-linux-arm64'
+    throw new Error(`Unsupported Linux architecture: ${process.arch}`)
+  }
+  throw new Error(`Python download is not supported on ${process.platform}`)
+}
+
 function getArchiveBase(): string {
   // LTX_PYTHON_URL is a dev-only override for testing with local archives.
   // Disabled in production to prevent code injection into a signed app.
@@ -255,7 +266,8 @@ function getArchiveBase(): string {
 function getFallbackArchiveUrl(): string | null {
   const hash = readHash(getBundledHashPath())
   if (!hash) return null
-  return `${FALLBACK_CDN_BASE}/python-embed-win32/${hash}/python-embed-win32.tar.gz`
+  const prefix = getPythonArchivePrefix()
+  return `${FALLBACK_CDN_BASE}/${prefix}/${hash}/${prefix}.tar.gz`
 }
 
 function isLocalPath(source: string): boolean {
@@ -314,7 +326,7 @@ export async function downloadPythonEmbed(
 ): Promise<void> {
   const destDir = path.join(app.getPath('userData'), 'python')
   const tempDir = path.join(app.getPath('userData'), 'python-tmp')
-  const archivePath = path.join(app.getPath('userData'), 'python-embed-win32.tar.gz')
+  const archivePath = path.join(app.getPath('userData'), `${getPythonArchivePrefix()}.tar.gz`)
 
   try {
     if (fs.existsSync(tempDir)) {
@@ -395,7 +407,7 @@ async function acquirePartsLocal(
   cleanupFiles: string[],
   onProgress: (progress: PythonSetupProgress) => void
 ): Promise<void> {
-  const manifestPath = path.join(dirPath, 'python-embed-win32.manifest.json')
+  const manifestPath = path.join(dirPath, `${getPythonArchivePrefix()}.manifest.json`)
   const manifest: ArchiveManifest = JSON.parse(fs.readFileSync(manifestPath, 'utf-8'))
 
   const partPaths: string[] = []
@@ -423,8 +435,9 @@ async function acquirePartsRemote(
   onProgress: (progress: PythonSetupProgress) => void
 ): Promise<void> {
   // Fetch manifest
-  const manifestUrl = `${baseUrl}/python-embed-win32.manifest.json`
-  const manifestDest = path.join(app.getPath('userData'), 'python-embed-win32.manifest.json')
+  const prefix = getPythonArchivePrefix()
+  const manifestUrl = `${baseUrl}/${prefix}.manifest.json`
+  const manifestDest = path.join(app.getPath('userData'), `${prefix}.manifest.json`)
   cleanupFiles.push(manifestDest)
   await downloadFileRaw(manifestUrl, manifestDest)
   const manifest: ArchiveManifest = JSON.parse(fs.readFileSync(manifestDest, 'utf-8'))
diff --git a/electron/updater.ts b/electron/updater.ts
index b3f33aaf..8caeb293 100644
--- a/electron/updater.ts
+++ b/electron/updater.ts
@@ -13,20 +13,20 @@ export function initAutoUpdater(
     autoUpdater.allowPrerelease = true
   }
 
-  // On Windows, don't auto-install — we need to pre-download python-embed first.
+  // On Windows/Linux, don't auto-install — we need to pre-download python-embed first.
   // On macOS, python is bundled in the DMG so auto-install is fine.
-  if (process.platform === 'win32') {
+  if (process.platform !== 'darwin') {
     autoUpdater.autoInstallOnAppQuit = false
   }
 
   autoUpdater.on('update-downloaded', async (info: UpdateDownloadedEvent) => {
-    if (process.platform !== 'win32') {
+    if (process.platform === 'darwin') {
       // macOS: python is bundled, just install normally
       autoUpdater.quitAndInstall(false, true)
       return
     }
 
-    // Windows: pre-download python-embed if deps changed before restarting
+    // Windows/Linux: pre-download python-embed if deps changed before restarting
     const newVersion = info.version
     logger.info( `[updater] Update downloaded: v${newVersion}, checking python deps...`)
 
diff --git a/frontend/views/editor/ClipContextMenu.tsx b/frontend/views/editor/ClipContextMenu.tsx
index 0005158e..010c6dcc 100644
--- a/frontend/views/editor/ClipContextMenu.tsx
+++ b/frontend/views/editor/ClipContextMenu.tsx
@@ -586,7 +586,9 @@ function SingleClipMenu({
               filePath = liveAsset.takes[Math.max(0, Math.min(takeIdx, liveAsset.takes.length - 1))].path
             }
             if (!filePath) return null
-            const label = window.electronAPI?.platform === 'darwin' ? 'Reveal in Finder' : 'Show in Explorer'
+            const label = window.electronAPI?.platform === 'darwin' ? 'Reveal in Finder'
+              : window.electronAPI?.platform === 'linux' ? 'Show in Files'
+              : 'Show in Explorer'
             return <MenuItem icon={FolderOpen} label={label} onClick={() => { window.electronAPI?.showItemInFolder(filePath); close() }} />
           })()}
         </>
diff --git a/package.json b/package.json
index 000eb816..0d7223fb 100644
--- a/package.json
+++ b/package.json
@@ -4,27 +4,26 @@
   "description": "LTX-2 Video Generation - Desktop App",
   "type": "module",
   "main": "dist-electron/main.js",
-  "author": "Lightricks",
+  "author": {
+    "name": "Lightricks",
+    "email": "ltx-desktop-dev@lightricks.com"
+  },
+  "homepage": "https://github.com/Lightricks/LTX-Desktop",
   "license": "Apache-2.0",
   "private": true,
   "packageManager": "pnpm@10.30.3",
   "scripts": {
-    "setup:dev:mac": "bash scripts/setup-dev.sh",
-    "setup:dev:win": "powershell -ExecutionPolicy Bypass -File scripts/setup-dev.ps1",
+    "setup:dev": "node scripts/run-script.js scripts/setup-dev",
     "dev": "vite",
     "dev:debug": "cross-env BACKEND_DEBUG=1 ELECTRON_DEBUG=1 vite",
     "typecheck": "concurrently \"pnpm run typecheck:ts\" \"pnpm run typecheck:py\"",
     "typecheck:ts": "tsc --noEmit",
     "typecheck:py": "cd backend && uv run pyright",
     "backend:test": "cd backend && uv sync --frozen --extra test --extra dev && uv run pytest -v --tb=short",
-    "build:mac": "bash scripts/local-build.sh --platform mac",
-    "build:win": "powershell -ExecutionPolicy Bypass -File scripts/local-build.ps1",
-    "build:mac:skip-python": "bash scripts/local-build.sh --platform mac --skip-python",
-    "build:win:skip-python": "powershell -ExecutionPolicy Bypass -File scripts/local-build.ps1 -SkipPython",
-    "build:fast:mac": "bash scripts/local-build.sh --platform mac --unpack --skip-python",
-    "build:fast:win": "powershell -ExecutionPolicy Bypass -File scripts/local-build.ps1 -Unpack -SkipPython",
-    "prepare:python:mac": "bash scripts/prepare-python.sh",
-    "prepare:python:win": "powershell -ExecutionPolicy Bypass -File scripts/prepare-python.ps1",
+    "build": "node scripts/run-script.js scripts/local-build",
+    "build:skip-python": "node scripts/run-script.js scripts/local-build --skip-python",
+    "build:fast": "node scripts/run-script.js scripts/local-build --unpack --skip-python",
+    "prepare:python": "node scripts/run-script.js scripts/prepare-python",
     "build:frontend": "vite build",
     "start:unpacked:mac": "open release/mac-arm64/LTX\\ Desktop.app",
     "start:unpacked:win": "powershell -Command \"& '.\\release\\win-unpacked\\LTX Desktop.exe'\""
diff --git a/resources/icons/1024x1024.png b/resources/icons/1024x1024.png
new file mode 100644
index 00000000..fa9c6bab
Binary files /dev/null and b/resources/icons/1024x1024.png differ
diff --git a/resources/icons/128x128.png b/resources/icons/128x128.png
new file mode 100644
index 00000000..d2ca80cf
Binary files /dev/null and b/resources/icons/128x128.png differ
diff --git a/resources/icons/16x16.png b/resources/icons/16x16.png
new file mode 100644
index 00000000..54cb7692
Binary files /dev/null and b/resources/icons/16x16.png differ
diff --git a/resources/icons/24x24.png b/resources/icons/24x24.png
new file mode 100644
index 00000000..132f4537
Binary files /dev/null and b/resources/icons/24x24.png differ
diff --git a/resources/icons/256x256.png b/resources/icons/256x256.png
new file mode 100644
index 00000000..6a6a5de7
Binary files /dev/null and b/resources/icons/256x256.png differ
diff --git a/resources/icons/32x32.png b/resources/icons/32x32.png
new file mode 100644
index 00000000..5e73f03d
Binary files /dev/null and b/resources/icons/32x32.png differ
diff --git a/resources/icons/48x48.png b/resources/icons/48x48.png
new file mode 100644
index 00000000..00860d85
Binary files /dev/null and b/resources/icons/48x48.png differ
diff --git a/resources/icons/512x512.png b/resources/icons/512x512.png
new file mode 100644
index 00000000..e8cfd537
Binary files /dev/null and b/resources/icons/512x512.png differ
diff --git a/resources/icons/64x64.png b/resources/icons/64x64.png
new file mode 100644
index 00000000..2b7cf223
Binary files /dev/null and b/resources/icons/64x64.png differ
diff --git a/scripts/Dockerfile.linux-build b/scripts/Dockerfile.linux-build
new file mode 100644
index 00000000..b6a2f2d8
--- /dev/null
+++ b/scripts/Dockerfile.linux-build
@@ -0,0 +1,26 @@
+FROM node:24-bookworm
+
+# System deps for electron-builder (AppImage/deb) + Python build
+RUN apt-get update && apt-get install -y --no-install-recommends \
+    git curl ca-certificates \
+    # electron-builder needs these for AppImage/deb packaging
+    libarchive-tools rpm \
+    && rm -rf /var/lib/apt/lists/*
+
+# uv (Python package manager)
+COPY --from=ghcr.io/astral-sh/uv:latest /uv /usr/local/bin/uv
+
+# pnpm
+RUN corepack enable && corepack prepare pnpm@latest --activate
+
+WORKDIR /app
+
+# Layer 1: Node dependencies (cached unless package.json/lockfile change)
+COPY package.json pnpm-lock.yaml pnpm-workspace.yaml ./
+RUN pnpm install --frozen-lockfile
+
+# Layer 2: Everything else
+COPY . .
+
+# Build frontend + prepare Python env + package installer
+CMD ["bash", "-c", "bash scripts/prepare-python.sh && pnpm build:frontend && bash scripts/create-installer.sh --platform linux"]
diff --git a/scripts/create-installer.sh b/scripts/create-installer.sh
index 02310f09..28f6fe93 100755
--- a/scripts/create-installer.sh
+++ b/scripts/create-installer.sh
@@ -67,7 +67,7 @@ if [ ! -d "dist" ] || [ ! -d "dist-electron" ]; then
   exit 1
 fi
 
-if [ ! -d "python-embed" ]; then
+if [ "$PLATFORM" != "linux" ] && [ ! -d "python-embed" ]; then
   echo "ERROR: Python environment not found. Run local-build.sh or prepare-python.sh first."
   exit 1
 fi
@@ -114,6 +114,13 @@ if [ "$UNPACK" = true ]; then
       echo "Unpacked app ready!"
       echo "Run: $RELEASE_DIR/win-unpacked/LTX Desktop.exe"
       ;;
+    linux)
+      echo ""
+      echo "Unpacked app ready!"
+      LINUX_UNPACKED="$RELEASE_DIR/linux-unpacked"
+      [ -d "$RELEASE_DIR/linux-arm64-unpacked" ] && LINUX_UNPACKED="$RELEASE_DIR/linux-arm64-unpacked"
+      echo "Run: $LINUX_UNPACKED/ltx-desktop"
+      ;;
   esac
 else
   echo ""
diff --git a/scripts/prepare-python.sh b/scripts/prepare-python.sh
index 544bfacf..53b59b47 100755
--- a/scripts/prepare-python.sh
+++ b/scripts/prepare-python.sh
@@ -1,12 +1,12 @@
 #!/usr/bin/env bash
 # prepare-python.sh
-# Downloads a standalone Python and installs all dependencies for macOS distribution.
+# Downloads a standalone Python and installs all dependencies for macOS/Linux distribution.
 #
 # Dependencies are read from uv.lock (via `uv export`) — pyproject.toml is the
 # single source of truth. No hardcoded dependency lists.
 #
 # Uses python-build-standalone (https://github.com/astral-sh/python-build-standalone)
-# which provides relocatable Python builds for macOS.
+# which provides relocatable Python builds for macOS and Linux.
 #
 # Prerequisites:
 #   - uv must be installed (https://docs.astral.sh/uv/)
@@ -36,11 +36,20 @@ case "$ARCH" in
   *) echo "ERROR: Unsupported architecture: $ARCH"; exit 1 ;;
 esac
 
-PBS_URL="https://github.com/astral-sh/python-build-standalone/releases/download/${PBS_TAG}/cpython-${PYTHON_VERSION}+${PBS_TAG}-${PBS_ARCH}-apple-darwin-install_only_stripped.tar.gz"
+# Detect OS for python-build-standalone target triple
+case "$(uname -s)" in
+  Darwin) PBS_OS="apple-darwin" ;;
+  Linux)  PBS_OS="unknown-linux-gnu" ;;
+  *)      echo "ERROR: Unsupported OS: $(uname -s)"; exit 1 ;;
+esac
+
+PBS_URL="https://github.com/astral-sh/python-build-standalone/releases/download/${PBS_TAG}/cpython-${PYTHON_VERSION}+${PBS_TAG}-${PBS_ARCH}-${PBS_OS}-install_only_stripped.tar.gz"
+
+PLATFORM_LABEL="$(uname -s) ($ARCH)"
 
 echo "========================================"
 echo "  LTX Video - Python Environment Setup"
-echo "  Platform: macOS ($ARCH)"
+echo "  Platform: $PLATFORM_LABEL"
 echo "  Python: $PYTHON_VERSION"
 echo "========================================"
 
@@ -77,8 +86,8 @@ echo "Step 2: Generating requirements.txt from uv.lock..."
 REQUIREMENTS_FILE="$BACKEND_DIR/requirements-dist.txt"
 
 # Export pinned deps, excluding the project itself.
-# Running on macOS auto-excludes Windows-only deps (triton-windows, pynvml, sageattention)
-# via sys_platform markers in pyproject.toml.
+# Platform markers in pyproject.toml auto-exclude irrelevant deps
+# (e.g. triton-windows on Linux/macOS, triton on macOS/Windows).
 uv export --frozen --no-hashes --no-editable --no-emit-project \
     --no-header --no-annotate \
     --project "$BACKEND_DIR" \
@@ -150,9 +159,14 @@ echo ""
 echo "Step 6: Installing dependencies from requirements.txt..."
 echo "  (This may take a while — PyTorch + ML libraries are large)"
 
-# No --extra-index-url needed on macOS: standard PyPI torch includes MPS support
+PIP_EXTRA_ARGS=()
+if [ "$PBS_OS" = "unknown-linux-gnu" ]; then
+  # Linux needs CUDA PyTorch wheels from the PyTorch index
+  PIP_EXTRA_ARGS+=(--extra-index-url "https://download.pytorch.org/whl/cu128")
+fi
+# On macOS, no --extra-index-url needed: standard PyPI torch includes MPS support
 "$PYTHON_EXE" -m pip install -r "$REQUIREMENTS_FILE" \
-    --no-warn-script-location --quiet
+    --no-warn-script-location --quiet "${PIP_EXTRA_ARGS[@]+"${PIP_EXTRA_ARGS[@]}"}"
 
 echo "  All dependencies installed"
 
@@ -178,17 +192,22 @@ find "$OUTPUT_PATH/lib" -type d -name "test" -exec rm -rf {} + 2>/dev/null || tr
 
 # Remove files only needed for building native extensions, not at runtime.
 # This cuts ~14k files and speeds up macOS codesigning dramatically.
-# NOTE: Windows needs .h files for sageattention/triton — this script is macOS only.
-rm -rf "$OUTPUT_PATH/include" "$OUTPUT_PATH/share" 2>/dev/null || true
-find "$OUTPUT_PATH/lib" -type d -name "include" -exec rm -rf {} + 2>/dev/null || true
+# NOTE: Linux needs .h files for sageattention/triton JIT compilation.
+if [ "$PBS_OS" = "apple-darwin" ]; then
+  rm -rf "$OUTPUT_PATH/include" "$OUTPUT_PATH/share" 2>/dev/null || true
+  find "$OUTPUT_PATH/lib" -type d -name "include" -exec rm -rf {} + 2>/dev/null || true
+  find "$OUTPUT_PATH" -name "*.h" -delete 2>/dev/null || true
+  find "$OUTPUT_PATH" -name "*.cuh" -delete 2>/dev/null || true
+  find "$OUTPUT_PATH" -name "*.cu" -delete 2>/dev/null || true
+else
+  # Linux: keep .h/.cuh/.cu files for triton/sageattention JIT, but remove other build artifacts
+  rm -rf "$OUTPUT_PATH/share" 2>/dev/null || true
+fi
 find "$OUTPUT_PATH" -name "*.pyi" -delete 2>/dev/null || true
 find "$OUTPUT_PATH" -name "*.pxd" -delete 2>/dev/null || true
 find "$OUTPUT_PATH" -name "*.pyx" -delete 2>/dev/null || true
 find "$OUTPUT_PATH" -name "*.hpp" -delete 2>/dev/null || true
 find "$OUTPUT_PATH" -name "*.cpp" -delete 2>/dev/null || true
-find "$OUTPUT_PATH" -name "*.h" -delete 2>/dev/null || true
-find "$OUTPUT_PATH" -name "*.cuh" -delete 2>/dev/null || true
-find "$OUTPUT_PATH" -name "*.cu" -delete 2>/dev/null || true
 find "$OUTPUT_PATH" -name "*.cmake" -delete 2>/dev/null || true
 
 # Remove temp directory and generated requirements file
@@ -205,12 +224,19 @@ echo "Step 8: Verifying installation..."
 
 "$PYTHON_EXE" -c "
 import sys
+import platform
 print(f'  Python: {sys.version}')
 try:
     import torch
     print(f'  PyTorch: {torch.__version__}')
-    mps = hasattr(torch.backends, 'mps') and torch.backends.mps.is_available()
-    print(f'  MPS available: {mps}')
+    if platform.system() == 'Darwin':
+        mps = hasattr(torch.backends, 'mps') and torch.backends.mps.is_available()
+        print(f'  MPS available: {mps}')
+    elif platform.system() == 'Linux':
+        cuda = torch.cuda.is_available()
+        print(f'  CUDA available: {cuda}')
+        if cuda:
+            print(f'  CUDA version: {torch.version.cuda}')
 except ImportError as e:
     print(f'  PyTorch import FAILED: {e}')
     sys.exit(1)
@@ -236,7 +262,7 @@ except ImportError as e:
 
 # Calculate size
 SIZE_BYTES=$(du -sb "$OUTPUT_PATH" 2>/dev/null | cut -f1 || du -sk "$OUTPUT_PATH" | awk '{print $1 * 1024}')
-SIZE_GB=$(echo "scale=2; $SIZE_BYTES / 1073741824" | bc)
+SIZE_GB=$(awk "BEGIN {printf \"%.2f\", $SIZE_BYTES / 1073741824}")
 
 echo ""
 echo "========================================"
diff --git a/scripts/run-script.js b/scripts/run-script.js
new file mode 100644
index 00000000..85b7c931
--- /dev/null
+++ b/scripts/run-script.js
@@ -0,0 +1,34 @@
+#!/usr/bin/env node
+// Runs a bash script on macOS/Linux or a PowerShell script on Windows.
+// Usage: node scripts/run-script.js <base-name> [args...]
+//
+// Example: node scripts/run-script.js scripts/local-build --skip-python
+//   macOS/Linux → bash scripts/local-build.sh --skip-python
+//   Windows     → powershell -ExecutionPolicy Bypass -File scripts/local-build.ps1 -SkipPython
+//
+// Arg conversion for PowerShell: --foo-bar → -FooBar, --foo → -Foo
+
+import { execSync } from 'child_process'
+
+let [,, baseName, ...args] = process.argv
+
+if (!baseName) {
+  console.error('Usage: node scripts/run-script.js <script-base> [args...]')
+  process.exit(1)
+}
+
+// Strip platform-specific extensions so tab-completed paths work
+baseName = baseName.replace(/\.(sh|ps1)$/, '')
+
+if (process.platform === 'win32') {
+  // Convert --kebab-case args to -PascalCase for PowerShell
+  const psArgs = args.map(arg => {
+    if (!arg.startsWith('--')) return arg
+    return '-' + arg.slice(2).split('-').map(w => w[0].toUpperCase() + w.slice(1)).join('')
+  })
+  const cmd = `powershell -ExecutionPolicy Bypass -File ${baseName}.ps1 ${psArgs.join(' ')}`
+  execSync(cmd, { stdio: 'inherit' })
+} else {
+  const cmd = `bash ${baseName}.sh ${args.join(' ')}`
+  execSync(cmd, { stdio: 'inherit' })
+}
diff --git a/scripts/setup-dev.sh b/scripts/setup-dev.sh
index 46c7f070..476a6d8f 100755
--- a/scripts/setup-dev.sh
+++ b/scripts/setup-dev.sh
@@ -1,5 +1,5 @@
 #!/usr/bin/env bash
-# macOS development setup for LTX Desktop
+# macOS / Linux development setup for LTX Desktop
 set -euo pipefail
 
 RED='\033[0;31m'
@@ -34,17 +34,26 @@ cd "$PROJECT_DIR/backend"
 uv sync --extra dev
 ok "uv sync complete"
 
-# Verify torch + MPS
+# Verify torch + accelerator
 echo ""
-echo "Verifying PyTorch MPS support..."
-.venv/bin/python -c "import torch; mps=hasattr(torch.backends,'mps') and torch.backends.mps.is_available(); print(f'MPS available: {mps}')" || true
+if [ "$(uname -s)" = "Darwin" ]; then
+  echo "Verifying PyTorch MPS support..."
+  .venv/bin/python -c "import torch; mps=hasattr(torch.backends,'mps') and torch.backends.mps.is_available(); print(f'MPS available: {mps}')" || true
+else
+  echo "Verifying PyTorch CUDA support..."
+  .venv/bin/python -c "import torch; cuda=torch.cuda.is_available(); print(f'CUDA available: {cuda}')" || true
+fi
 
 # ── ffmpeg check ────────────────────────────────────────────────────
 echo ""
 if command -v ffmpeg >/dev/null 2>&1; then
   ok "ffmpeg found: $(ffmpeg -version 2>&1 | head -1)"
 else
-  echo "⚠  ffmpeg not found — install with: brew install ffmpeg"
+  if [ "$(uname -s)" = "Darwin" ]; then
+    echo "⚠  ffmpeg not found — install with: brew install ffmpeg"
+  else
+    echo "⚠  ffmpeg not found — install with: sudo apt install ffmpeg  (or sudo dnf install ffmpeg)"
+  fi
   echo "   (imageio-ffmpeg bundled binary will be used as fallback)"
 fi