Arcads is an AI-powered content creation platform that lets you generate product images, talking actor videos, B-roll clips, and full ad campaigns — without a studio, crew, or production budget.

This guide covers every feature available on the platform so you always know what's possible and where to find it.

Tools are standalone utilities you can apply to any video or image you've generated — or use independently. Find them under See More on your dashboard.

Automatically generate and overlay subtitles onto any video. Captions are synced to the audio and styleable to match your brand. Ideal for social content watched without sound.

👉 Open Add Captions

Convert the spoken audio from any video into a text transcript. Useful for reviewing script delivery, creating manual subtitles, or repurposing video content into written formats.

👉 Open Transcribe

Localize any video into a different language automatically — including lip sync matched to the new audio. Select your target language, choose an accent if available, and Arcads generates the translated version. No re-filming required.

👉 Open Translate Video

Generate a voiceover from a written script using Arcads' AI voice library. Choose from dozens of voices across multiple languages and accents, or use a cloned voice for brand consistency.

👉 Open Text to Speech

Transform an existing audio recording into a different AI voice while preserving the timing and delivery pattern. Useful for voice swaps without re-recording the script.

👉 Open Speech to Speech

Replace the voice in an existing video with a different AI voice. Useful for swapping accents, gender, or tone on already-generated videos without touching the visuals.

👉 Open Change Voice

Speed up or slow down any video clip. Helpful for trimming duration to fit ad platform requirements (e.g., cutting a 20s video down to a 15s format).

👉 Open Change Speed

Instantly remove the background from any image or video. The result is a clean cutout ready to be placed on a custom background or used in compositing — no manual masking needed.

👉 Open Remove Background

Improves the realism and quality of AI-generated faces. Adds natural skin detail and reduces the "AI look" — particularly useful for close-up talking-head content.

👉 Open Skin Enhancer

Increase the resolution of any video using AI. Useful when your source footage is lower quality and needs to meet platform resolution requirements before publishing.

👉 Open Upscale

Found a proven viral hook? Feed it in with your product details and the tool recreates it tailored to your brand. Great for testing multiple opening angles on a single script.

👉 Open Hook Repurposer

Add text graphics on top of any video — titles, captions, CTAs, or branding. Customizable font, position, and timing.

👉 Open Text Overlay

Cut a video to a specific time range to isolate the best portion of a clip. Standard non-destructive editing — original file is preserved.

👉 Open Trim Video

Change the aspect ratio or dimensions of any video (e.g., 16:9 → 9:16 for Reels/TikTok, 1:1 for feed). Essential for multi-platform distribution from a single generation.

👉 Open Resize Video

Combine multiple video or image layers into a single composited output. Used for adding B-roll overlays, product shots, or graphic elements on top of actor footage.

👉 Open Merge Layers

Join multiple video clips sequentially into one continuous video. Use this to combine an intro hook, main body, and outro CTA into a single finished file.

👉 Open Stitch Videos

Lengthen any generated video by adding seamless continuation frames using AI. Useful when a clip is just slightly too short for your target duration.

👉 Open Extend Video

Pull a single still frame from any video. Great for generating thumbnails, using a specific moment as a reference image, or creating static ads from a video.

👉 Open Extract Frame

Adjust or simulate a different camera angle on an existing image or video using AI. Repositions the shot without re-generating the whole scene from scratch.

👉 Open Camera Angle

A full AI-powered video editing interface driven by the Kling o3 model. Describe what you want changed in plain language — add, remove, or replace elements in a video — and the model applies the edit.

👉 Open Video Edit (Kling o3)

AI video editing powered by xAI's Grok model. An alternative to Kling o3 for instruction-based edits, offering a different visual style and output characteristic.

👉 Open Video Edit (Grok)

Takes a static image of an actor and animates it into a short video using the Wan 2.2 model. Gives life to still photos with natural-looking movement — without a full Talking Actor pipeline.

👉 Open Animate Actor (Wan 2.2)

Animates an actor image using Kling 2.6 Pro with specific, directed motion control parameters. More precise than standard animation — useful when you want the actor to move in a particular way.

👉 Open Kling 2.6 Motion Control

The highest-quality motion control option for actor animation. Powered by Kling 3.0 Pro — best for premium outputs that need highly realistic, directed movement.

👉 Open Kling 3.0 Motion Control

Visualize any clothing item on an AI model without production, photography, or physical samples. Upload the garment and Arcads places it on a model of your choice. Ideal for fashion and apparel brands.

👉 Learn about Fashion Try-On

Automatically detect scene changes and split a video into separate clips.
👉 Open Split Into Scenes

Generate original music from a text prompt with adjustable duration and instrumental mode.

👉 Open Create Music

​

Arcads integrates multiple leading AI image generation models. Each has different strengths — choose based on your visual style, use case, and desired output quality.

OpenAI's image generation model integrated into Arcads. Reliable for generating product shots, lifestyle backgrounds, and general ad visuals from text prompts.

👉 Open GPT Image 1.5

The newer, higher-quality version of OpenAI's image model. Produces sharper, more detailed, and more instruction-following outputs compared to GPT Image 1.5.

👉 Open GPT Image 2

Arcads' own proprietary image generation model. Fast and optimized for UGC-style ad imagery — great for quick iterations and high-volume image workflows.

👉 Open Nano-Banana Pro

The updated version of Nano-Banana with improved realism and visual consistency across generations. A step up in quality for the same fast, UGC-optimized workflow.

👉 Open Nano-Banana 2

A high-quality image generation model with strong creative and stylized output capabilities. Well-suited for editorial, aspirational, and lifestyle aesthetics.

👉 Open Seedream 4.5

A faster, lighter variant of Seedream 5. Balances quality and speed — ideal for higher-volume workflows where you need good results quickly.

👉 Open Seedream 5 Lite

Image generation powered by xAI's Grok model. Offers a distinct visual style and output aesthetic compared to GPT or Seedream — useful for variety or when other models don't quite hit the look you want.

👉 Open Grok Image

A specialized image tool powered by the Soul engine. Generates authentic UGC-style images — selfie-aesthetic, natural lighting, lifestyle product contexts — rather than polished studio shots. Best for brands that want human, creator-feel imagery.

👉 Learn about UGC Studio

These models generate B-roll clips, cinematic product shots, and movement-based video from your images or text prompts. Use them independently or combine with the Talking Actor pipeline for full ad productions.

OpenAI's Sora video generation model, integrated directly into Arcads. Generates high-quality videos from text prompts with strong scene consistency and photorealistic output.

👉 Open Sora 2

The Pro variant of Sora 2. Delivers longer duration, higher resolution outputs, and more cinematic quality — best for hero shots and premium ad placements.

👉 Open Sora 2 Pro

Google DeepMind's video generation model. Produces cinematic, high-fidelity video from text or image prompts. Known for strong temporal consistency and highly realistic motion — particularly good for lifestyle and environmental scenes.

👉 Open Veo 3.1

A leading video generation model known for realistic motion and photorealism. A strong choice for lifestyle and product-focused video ads.

👉 Open Kling 2.6 Pro

The flagship Kling model. Highest quality output in the Kling line — improved motion realism, better scene coherence, and more cinematic results. Recommended for publication-ready B-roll and product content.

👉 Open Kling 3.0 Pro

Kling 3.0 at 4K resolution. For content that requires the highest possible visual fidelity — premium ad placements, connected TV, or large-format display.

👉 Open Kling 3.0 4K

Generate fast 1080p videos from text or a starting image with Kling 3 Pro Turbo.

👉 Open Kling 3 Pro Turbo

Optimized for fluid, dynamic video generation. Particularly good for product movement, lifestyle scenes, and short-form ad content. A reliable workhorse for B-roll production.

👉 Open Seedance 1.5

The updated Seedance model with improved motion quality and better handling of complex scenes. Use this over 1.5 for final outputs.

👉 Open Seedance 2.0

A specialized video generation model within Arcads optimized for stylized and animated-style content. Best for creative, non-photorealistic outputs where you want a distinct visual character.

👉 Happy Horse Prompting Guide

xAI's Grok-powered video generation. An alternative model option for users who want a different output style and aesthetic from the rest of the lineup.

👉 Open Grok Video

Talking Actor models animate an AI character to deliver a spoken script on camera — lip-synced to the audio, with expressions and movement. These are the core of Arcads' ad creation pipeline.

The original Arcads talking head model. Reliable and credit-efficient — best for volume production where speed and cost matter more than maximum realism. Good starting point for testing scripts before committing to higher-quality renders.

👉 Open Arcads 1.0
​

Generates a talking actor video where lip sync is driven by a provided audio file rather than a text script. Ideal when you already have a recorded voiceover and want to attach it to an actor without re-generating audio.

👉 Open Audio Driven
​

The most advanced Talking Actor model on Arcads. Produces highly realistic full-body motion, gestures, and facial expressions — not just lip sync. Closest to real human video. Best for polished, publication-ready ads.

👉 Open OmniHuman 1.5
​

💡 Not sure which model to use? Compare all three Talking Actor models here.

Browse a diverse pre-built library of AI actors across ages, ethnicities, styles, and settings. Any actor can be animated with the Talking Actor models and filtered by category to find the right fit for your brand.

Create your own AI actor from a single headshot image. Upload a photo, and Arcads generates an animatable actor in your chosen style — saved to your account for reuse across multiple videos.

👉 How to Create a Custom Talking Actor

Create a hyper-realistic digital twin of yourself — replicating both your face and your voice. Once cloned, generate unlimited videos without re-recording anything.

👉 How to Clone Yourself Into a Talking Actor

Choose from a library of AI voices across multiple languages and accents. Pro users can also clone any voice — upload a short audio sample and Arcads replicates it for consistent use across all your content.

Add expressive hand movements and body language to any AI actor using preset gesture templates (thumbs up, pointing, open hands, and more). Makes actors appear more dynamic and engaging on screen.

👉 Learn about Gestures

Import any product image into Arcads as the base for image generation, video creation, and product showcase content. Arcads detects the product automatically — no manual cropping required.

Place your product directly into any generated video — held by an actor, displayed on a surface, or shown in a lifestyle context.

👉 How to Show a Product in a Video

Generate content showing your app's UI alongside a talking actor or voiceover — no screen recording software or editing skills needed. Built for SaaS and mobile app companies.

👉 How to Show Your Mobile App in a Video

Generate unboxing-style product reveal videos from your product image — no physical box or filming required. Ideal for e-commerce product launch campaigns.

👉 How to Create Unboxing Videos

Swap the actor in an existing video with a different one while preserving the script, timing, and background. Great for A/B testing the same creative with different faces.

👉 How to Replace an Actor in a Video

Presets are pre-configured workflows that combine multiple Arcads features into a single, guided creation flow. They're the fastest way to produce a specific type of content without configuring each step manually.

Combines Sora 2 video generation with AI actors in a pre-configured flow. Simplifies the process of creating high-quality actor-led video ads using Sora's capabilities.

👉 Open Sora 2 Actors

Generates a first-person, point-of-view unboxing video. Great for e-commerce brands wanting product reveal content that feels authentic without physical filming.

👉 Learn about Unboxing Videos

A preset focused on highlighting a product with close-ups, angles, and feature callouts. Optimized for e-commerce and DTC product ads where the product is the hero.

👉 Open Product Showcase

Adds simulated camera motion (pan, zoom, dolly) to a static or generated video. Makes content feel more cinematic and professionally shot.

👉 Open Camera Movement

A preset for generating mobile game advertising content. Typically overlays gameplay footage with a talking actor or hook-style intro to drive installs.

👉 Open Gameplay Ad

Workflows let you connect multiple Arcads features into a single automated pipeline. Build once, generate at scale — instead of creating assets one by one, you run the whole sequence automatically.

What you can automate:

How it works:

Pre-built workflow templates are available for the most common use cases. Open one, swap in your assets, and run.

👉 Learn more about Workflows

The listed features above consume credits when you generate content. Credit usage varies by model — higher-quality models (Kling 3.0 Pro, Sora 2 Pro, etc.) use more credits per generation than faster ones.

Image generation is free (up to 100 per day). Video and Talking Actor generations consume credits, too.

👉 How Are Credits Counted?
👉 How to Buy More Credits

Arcads, by default, supports content creation across a wide range of languages. It's been beefed up recently via the Translate Video and Text to Speech features. Pro accounts have additional voice customization with the help of ElevenLabs integration.

👉 What Languages Are Supported?