Veo Automation: VEO & Nano Banana Google Flow

Veo Automation for Google Flow — free bulk VEO 3 & Nano Banana generation, auto-download 4K, batch prompt queue at scale.

The only FREE extension that automates Google Flow completely. Generate hundreds of VEO 3 videos and Nano Banana images in batch — hands-free. Auto-download in 4K. No subscription required. ──────────────────────────────────────── 🎬 WHAT IS VEO AUTOMATION FREE? ──────────────────────────────────────── VEO Automation Free is a Chrome extension that works directly inside Google Flow (labs.google/fx/tools/flow). It automates the entire generation workflow so you can queue up dozens or hundreds of prompts and walk away. When it finishes, every video or image is already downloaded to your computer. ──────────────────────────────────────── 🚀 KEY FEATURES ──────────────────────────────────────── ✅ FREE — All core features are completely free. No hidden paywalls. 🎬 TEXT TO VIDEO (VEO 3, VEO 3.1 Lite/Fast/Quality, Omni Flash) Batch-generate videos from a list of text prompts. Supports all Google Flow video models including Veo 3.1 Lite [Lower Priority] for Ultra plan users. 🖼 TEXT TO IMAGE (Nano Banana 2, Nano Banana Pro) Bulk-generate images from text prompts using all image models. 🖼→🎬 FRAME TO VIDEO Upload a reference image as a start frame (or start + end frame) and generate videos from it in batch. 🥗 INGREDIENTS TO VIDEO (Components to Video) Upload multiple ingredient images and generate videos that combine them. Control exactly how many ingredients are used per prompt (1–6). 🖼→🖼 IMAGE TO IMAGE Use reference images to guide image generation in batch mode. 🤖 AGENT MODE Automate Google Flow's built-in Agent mode. Let the AI decide composition, style, and generation parameters — just provide your prompts and Agent mode handles the rest in batch. 🌐 20 INTERFACE LANGUAGES Full UI in English, Spanish, Portuguese, French, German, Japanese, Chinese, Hindi, Arabic, Korean, Urdu, Turkish, Russian, Italian, Vietnamese, Indonesian, Bengali, Polish, Persian, and Dutch. Arabic, Urdu and Persian auto-switch to right-to-left layout. ──────────────────────────────────────── ⚡ AUTOMATION FEATURES ──────────────────────────────────────── 📋 Batch Prompt Queue • Paste any number of prompts (separated by blank lines) • Upload prompts from .txt files or Excel/CSV spreadsheets • Live queue shows each prompt's status: Queued → Reviewing → Generating (%) → Completed ⬇️ Auto-Download in 4K • Automatically downloads every generated video or image • Video quality: 720p (original), 1080p upscaled, 4K upscaled, Animated GIF • Image quality: 1K, 2K, 4K • Saves to a custom subfolder inside your Downloads directory • Auto-renames files with prompt index + prompt text for easy organisation ⏸ Pause & Resume • Pause a running batch at any point — automation finishes the current prompt then holds • Resume instantly and pick up exactly where you left off 🛡️ Smart Unusual Activity Protection • Detects when Google flags unusual activity and automatically cools down before retrying • Human-like review pause proportional to prompt length — short prompts pause briefly, long prompts pause longer • Live countdown visible in the prompt queue and a dismissible toast notification • Respects your retry limit — skips cooldown on the final retry attempt 🖼 Rich Image Upload UI • Drag-and-drop image upload with thumbnail previews • Drag to reorder images • Per-prompt image preview — see exactly which image pairs with which prompt 📋 Debug Logs with Copy • Live action log shows every step of the automation • Copy all logs to clipboard with one click for easy troubleshooting ⚙️ Flexible Output Settings • Aspect ratio: 16:9, 9:16, 4:3, 3:4, 1:1 • Outputs per prompt: 1–4 • Random delay between prompts (min–max seconds) • Max retries per prompt (auto-retries on failure, 0–10) • Concurrent prompt processing 🌓 Light & Dark Mode • Toggle between dark (default) and light themes instantly • Preference saved across sessions 🔍 Panel Size Controls • Resize the control panel to small, medium, or large • More screen space for your prompts and queue 🪟 Pop-Out Window Open the control panel in a floating window while Google Flow stays in the main tab. Freely resize and reposition. ──────────────────────────────────────── 🔒 PRIVACY & SAFETY ──────────────────────────────────────── • Works 100% locally — no data is sent to any server • Does NOT read, store, or transmit your prompts, videos, or images • Only activates on labs.google/fx/tools/flow — no access to any other website • Chrome Debugger API is used only to simulate trusted click events (required by React's event system); it never reads any page data • All settings stored locally in your browser only • Full privacy policy: https://veoautomation.com/privacy ──────────────────────────────────────── 📖 HOW TO USE ──────────────────────────────────────── 1. Install the extension and pin it to your Chrome toolbar 2. Navigate to https://labs.google/fx/tools/flow 3. The side panel opens automatically (or click the extension icon) 4. Choose your generation mode (Text to Video, Frame to Video, etc.) 5. Paste your prompts — one prompt per block, blank lines between them 6. Set your output settings (model, quality, aspect ratio, download folder) 7. Click ▶ Run and let Veo Automation handle the rest! Use Pause/Resume to stop and continue at any time. ──────────────────────────────────────── 📋 CHANGELOG ──────────────────────────────────────── v1.0.14 (Latest) FIX: Reference-image upload for Image to Image and Ingredients to Video is rebuilt again — the v1.0.13 mechanism worked reliably in testing but Chrome silently blocked it on some live installs. The new approach hands the image to Google Flow's own file picker directly, without relying on the browser's remote-debugging protocol at all, so it isn't affected by that restriction. v1.0.13 • NEW: Nano Banana 2 Lite — Google's newest image model is now available in the Model list for Text to Image and Image to Image, and is the new default. • FIX: Reference-image upload for Image to Image and Ingredients to Video was rebuilt on a more reliable mechanism after Chrome tightened a security check the previous approach relied on. Attaching a reference image is faster and more consistent now. • FIX: Auto-rename was silently skipping file numbering whenever "Save to Folder" was left blank — files kept Google's own name instead of your prompt numbering. Fixed for real this time; numbering now applies with or without a custom folder. • FIX: A prompt Google rejects for a content-policy reason now fails immediately and moves on to the next prompt, instead of retrying for several minutes. • NEW: When your Google Flow account runs out of video credits, the extension now recognizes it immediately, stops the run cleanly instead of retrying every remaining prompt, and the completion popup clearly shows "Out of Credits" with a link to refresh or upgrade. (Image generation is unaffected — it doesn't use credits.) • Frame to Video is temporarily disabled again while its reference-image upload step is rebuilt on the same new mechanism as Image to Image and Ingredients to Video — it will return in a follow-up release. v1.0.12 • FIX: Frame to Video, Image to Image and Ingredients to Video reference images stopped attaching after a recent Chrome update, failing with a "Not allowed" error. A stricter Chrome requirement for drag-and-drop automation is now satisfied, so reference images attach reliably again. • FIX: A prompt Google rejects for a content-policy reason (for example, content flagged as involving minors) used to sit retrying for a minute or more before finally being marked failed. It's now recognized immediately and the run moves straight to the next prompt. v1.0.11 • Frame to Video, Image to Image and Ingredients to Video are back — fully working on the new Google Flow. You add reference images only in the extension; it hands them to Flow for you, no manual upload needed. • How reference images reach Flow: the extension briefly saves each image to a "veo-automation-tmp" folder in your Downloads and deletes it automatically right after handing it over. A short note in the side panel explains the quick download flash you may see on the first prompt of an image run. • Stability fixes on flow.google.com: more reliable tracking of which new tile belongs to which prompt, video completion detection on the rebuilt tile UI, and the quality-download step-down (4K to 2K to 1K for images, 4K to 1080p to 720p for video) when an upscale is locked on your plan. • Agent mode stays disabled for now while it is updated for the new site — the button is visible but not selectable and shows a "being updated" note. v1.0.10 • MAJOR: Rebuilt for the new Google Flow (flow.google.com). Google moved Flow off labs.google/fx/tools/flow to flow.google.com and rebuilt the site, which broke the previous version. This release re-implements the automation for the new site. • Text to Video and Text to Image are fully working on flow.google.com — create project, configure model / aspect ratio / resolution / duration / output count, fill the prompt, generate, and auto-download every result. • Download quality with automatic fallback: images 4K -> 2K -> 1K, videos 4K -> 1080p -> 720p. If an upscale is locked on your plan it silently steps down instead of failing the prompt. Files are still numbered and saved to your chosen sub-folder. • Frame to Video, Ingredients to Video, Image to Image and Agent mode are temporarily marked "coming soon" while they are updated for the new site — the buttons are visible but not selectable. They return in a follow-up release. • The "Open Google Flow" button and in-panel instructions now point to flow.google.com. v1.0.9 • FIX: Image-to-Image, Frame-to-Video, and Ingredients-to-Video — reference images were lost after unusual activity triggered a session recovery. Every prompt now uploads the correct image after recovery without any user action. • FIX: Run stalling after recovery — after unusual activity opened a fresh session, the run would sometimes freeze with "heartbeat lost — reloading tab" and never continue. Recovery now completes reliably every time. • FIX: Prompt queue showing "Downloading..." during a retry — when a prompt was retrying after a partial result, the queue incorrectly showed Downloading instead of Retrying. Now shows the correct status. • FIX: Unusual activity cooldown stacking — after a cooldown, the retried prompt was being hit with both the cooldown wait and an additional 45–90s pacing delay (double penalty). The retry now correctly skips the extra pacing since it already waited. • FIX: Pacing delay applying to clean prompts — after unusual activity was resolved, the slow-down delay was incorrectly continuing into subsequent prompts that had no issues. Pacing now clears immediately after a successful generation. • IMPROVEMENT: Random delay added after image upload — a 5–10 second natural pause is inserted after each image upload before the prompt is submitted. Reduces unusual activity detection on image-based modes. Visible as "Random delay Xs..." in the prompt queue. • FIX: Prompt queue language — the live status chips (Configuring, Submitting, Locating, Retrying, Upscaling, Saving, Downloading, Reviewing, cooldown and upload labels) stayed in English after switching the interface language. They now follow the selected language across all 20 languages. • FIX: Full interface translation — the completion summary, rate-limit notice, new-project and terms-agreement dialogs, the unusual-activity toast, the Clear Cache button and several other labels were still hardcoded in English. Everything now switches with the chosen language. • FIX: Failed prompts marked as completed — when Google confirmed every output of a prompt as "Failed", the queue showed a green Completed instead of a red Failed. It now reports Failed with an accurate downloaded count of 0. • NEW: Partially Done status — when a prompt downloads some outputs but loses others to unusual activity, the queue now shows an amber "Partially Done" instead of a green Completed, so it is clear at a glance which prompts only produced part of their requested outputs. • IMPROVEMENT: Shorter slow-down after unusual activity — the protective pacing delay between prompts following an unusual-activity event was reduced from 45–90 seconds to 20–30 seconds, so runs recover faster while still easing off Google's rate limiter. v1.0.8 • NEW: Start # — set the starting number for downloaded file names. If you run multiple sessions, pick up numbering exactly where you left off (e.g. start at 51 → files named 51_prompt.mp4, 52…). Sits beside the Auto-rename toggle. Default: 1. • NEW: Omni Flash Duration — when Omni Flash model is selected, choose video duration (4s / 6s / 8s / 10s) directly in the extension. Selected duration is applied automatically in Google Flow. • FIX: Frame to Video 1080p upscaling — if 1080p upscaling fails, automatically retries once then falls back to 720p so no tile is ever skipped. • FIX: New Google Agent UI — detects and auto-closes the "What do you want to create?" prompt panel on page load before automation starts, restoring the normal generation UI automatically. • FIX: No Download mode — Completion dialog incorrectly showed "Downloaded: 10/10" instead of "Generated: 10". Stats now show the correct label and count for no-download runs. • FIX: Auto-rename OFF — files went to root Downloads folder instead of your chosen subfolder. Files now save to the correct subfolder with their original filenames. • FIX: Fresh session after unusual activity — a Google cookie consent banner was blocking tile detection after a fresh session was opened, causing the run to time out. Now auto-dismissed. • REMOVED: Imagen 4 — removed from image model options as it is no longer reliably available in Google Flow. v1.0.7 • Google Flow interface update — fixed image download issues caused by Google's latest changes • Fixed automation getting stuck — runs now continue automatically even if the page crashes or reloads • Better handling when Google blocks generations — the extension now slows down and retries in a fresh session automatically • New alert popup when Google limits your account — with simple tips to fix it (use 1 output per prompt) • Fixed Image-to-Image bulk runs freezing when using many reference images v1.0.6 • Image upscaling retry — if 2K/4K upscaling fails, automatically retries once; falls back to 1K so no image is ever skipped • Session stats fix — downloaded count now accurate when unusual activity recovery triggers mid-batch • Completion dialog always shows Close button — scrollable on short screens so it's always reachable v1.0.5 v1.0.4 v1.0.3 v1.0.2 v1.0.1 • Initial release — batch VEO & Nano Banana automation for Google Flow • 5 generation modes, auto-download 4K, 10 language UI ──────────────────────────────────────── 📧 SUPPORT ──────────────────────────────────────── Website: https://veoautomation.com Email: veosupport@devoveagency.com Telegram (immediate): @veoautomation Privacy Policy: https://veoautomation.com/privacy ---------------------------------------------------------------- WEBSITE / HOMEPAGE URL ---------------------------------------------------------------- https://veoautomation.com/ ---------------------------------------------------------------- PRIVACY POLICY URL (host this file at your domain) ---------------------------------------------------------------- https://veoautomation.com/privacy