Veo 3.1 & Imagen 3 API Ready

Cinematic AI Video,
Orchestrated at Scale

Supercharge your AI video production pipeline. Automate batch queues, map double-keyframe transitions, and synthesize synchronized background audio in a premium visual environment.

14,808+
Renders Orchestrated
100%
Storage Native Integration
Veo 3.1 Lite
Primary Model Support

The Production Challenge

Why AI Video Generation Stalls in standard workflows

01

The Scattered Pipeline

  • Manual Cloud Uploads: Manually converting and uploading keyframe starting images to storage is slow and breaks focus.
  • Disconnected Operations: Tracking long-running API operations manually results in massive downtime between clips.
  • Silent Visuals: AI models generate gorgeous clips, but they remain completely silent, requiring external audio splicing.
  • Fragile Prompts: Safety timeout blocks trigger warnings and stall entire queues without automated safety fallbacks.
02

The VeoFlow Studio System

  • Auto-Storage Sync: The client uploads frames to local buffers, uploads automatically to asset buckets, and kicks off generation.
  • Live Polling Hooks: Monitors the generation API operation status. Downloads output files automatically when complete.
  • Dual-Model Pipelines: Synthesize high-fidelity audio automatically inside the configuration payload for complete immersion.
  • Smart Safety Skip Countdowns: Configurable countdown warnings auto-skip offending prompts to keep queues moving.

Core Architecture

Designed around what the codebase actually does

Double-Keyframe Video Interpolation

Don't leave video motion to chance. Upload starting images and optional ending keyframe images (`inputImage` and `inputEndImage`). The studio automatically processes, compresses, and maps them to generate seamless, controlled transitions.

compressImage()uploadBase64ToStorage()

Synchronized Audio Engine

Inject atmospheric depth directly. Toggle audio generation in the dashboard config to call model parameters to generate matched soundscapes.

generateAudio: true

Queue Runner & Bulk Import

Type or paste lists of multiple prompts with our Bulk Import panel. Run, pause, or skip queues. The client tracks operation indices seamlessly.

Sandboxed Project Environments

Segment your historical records. Switch active environment IDs (`projectEnvId`) to isolate visual campaigns, prompts, and storage settings.

Interactive Safety Skip

If the safety filter flags a prompt due to guidelines, an interactive countdown modal prompts the operator. Automatically skip or halt queue to avoid quota lockouts.

Live Interactive Playground

Experience the rendering pipeline

Watch how Flow Video Studio orchestrates cloud storage, AI model calls, and real-time operation polling to generate cinematic content.

Video Generation Simulation
● remotion rendered15s · 450f
⚙️
Open settings
📐
Pick 9:16 ratio
🤖
Select model
✏️
Type prompt
🚀
Send & generate
🎬
Video renders

Pricing Options

Flexible tiers for creators and studios

Generations are billed dynamically by tokens based on the model selected.
(e.g. Veo 3.1 Lite costs 5 tokens/sec with audio, Imagen 3.0 costs 3 tokens/image).

Token Booster

Top Up Generation Credits

Need more generation capacity? Purchase additional tokens programmatically calculated at your active subscription plan rate.

Token Booster Locked

You do not currently have an active monthly subscription. You must purchase a Pro or Elite plan first to unlock add-on token purchasing.

Model Consumption & Plan Yields

Every generation consumes tokens dynamically based on the complexity and quality profile of the model. Video models are billed per generated second (including optional audio synchronization), while image models are billed per static asset.

Model Name & CapabilitiesToken RateHobbyist (Free)Pro Creator (15k Tokens)Elite Creator (30k Tokens)
🎬 Cinematic Video Generation (Veo Models)
Veo 3.1 Standard1080p Full HD output, premium physical motion simulation, and high cinematic photorealism.720p: 20 tokens/s (muted) | 40 tokens/s (audio)
1080p: 30 tokens/s (muted) | 50 tokens/s (audio)
Simulated PreviewWatermarked mock video300 to 750 secondse.g., ~60 to ~150 clips of 5s600 to 1,500 secondse.g., ~120 to ~300 clips of 5s
Veo 3.1 Fast720p/1080p, accelerated generation speeds. Perfect balance of speed and frame stability.720p: 10 tokens/s (muted) | 15 tokens/s (audio)
1080p: 15 tokens/s (muted) | 20 tokens/s (audio)
Simulated PreviewWatermarked mock video750 to 1,500 secondse.g., ~150 to ~300 clips of 5s1,500 to 3,000 secondse.g., ~300 to ~600 clips of 5s
Veo 3.1 Lite720p/1080p, optimized latency. Ideal for swift iterations and draft layout storyboarding.720p: 3 tokens/s (muted) | 5 tokens/s (audio)
1080p: 8 tokens/s (muted) | 10 tokens/s (audio)
Simulated PreviewWatermarked mock video1,500 to 5,000 secondse.g., ~300 to ~1,000 clips of 5s3,000 to 10,000 secondse.g., ~600 to ~2,000 clips of 5s
🖼️ High-Fidelity Image Synthesis (Imagen / Banana Models)
Nano Banana Pro (Imagen 3 Pro)Premium model configuration. Perfect for high-res advertising key visuals.Standard: 8 tokens / image
High-Res: 12 tokens / image
Simulated PreviewWatermarked SVG card1,250 to 1,875 imagesHigh resolution assets2,500 to 3,750 imagesHigh resolution assets
Nano Banana 2 (Imagen 3.1 Flash)Fast-rendering model configuration. Highly responsive storyboard layouts.Standard: 5 tokens / image
High-Res: 8 tokens / image
Simulated PreviewWatermarked SVG card1,875 to 3,000 imagesBalanced asset outputs3,750 to 6,000 imagesBalanced asset outputs
Nano Banana (Imagen 3.0 Standard)Standard text-to-image and image-to-image generator. Excellent general capability.Standard: 3 tokens / image
High-Res: 5 tokens / image
Simulated PreviewWatermarked SVG card3,000 to 5,000 imagesStandard asset outputs6,000 to 10,000 imagesStandard asset outputs
⚡ Interactive Chat Assistant (Gemini Models)
Gemini 3.5 Flash (Default Chat)Standard ultra-fast chat assistant. Ideal for prompt configuration advice.Dynamic (Google cost)
$1.50 In | $9.00 Out (per 1M)
2 Chat MessagesFlat 1 token / message~30k to ~150k messagesCalculated at $0.01 per token~60k to ~300k messagesCalculated at $0.01 per token
Gemini 3.1 Pro PreviewPremium complex reasoning chat assistant. Best for long-context scene scripting.Dynamic (Google cost)
$2.00 In | $12.00 Out (per 1M)
2 Chat MessagesFlat 1 token / message~22k to ~112k messagesCalculated at $0.01 per token~45k to ~225k messagesCalculated at $0.01 per token

Ready to scale your production?

Get early access to our hosted team features, centralized project workspaces, and advanced rendering queues. We're launching new regions soon.