Skip to content
SceneRender
What powers SceneRender

Many models, one place to make things.

SceneRender does not run on one AI. Each kind of result — a scene with recurring characters, a product shot, a presenter, a song, a line of narration — goes to the model that does that job best today, and the studios do the planning, checking and assembling around them. This page says which is which.

What you can create

Seven studios and an editor.

  • Clip Studio

    Multi-scene stories, skits, explainers and training videos: you describe it, review every scene, and assemble the final with narration, captions and music.

    Open Clip Studio

  • Video Studio

    One standalone AI video from a line of text or a reference image, in the shape and length you pick.

    Open Video Studio

  • Image Studio

    A picture from a description, or from a photo you have: thumbnails, pins, posts, characters, product scenes.

    Open Image Studio

  • Music Studio

    Original music from your words: a hook, a bed or a whole song, sung or instrumental, yours to use in what you post.

    Open Music Studio

  • Ad Studio

    Listing images, A+ tiles, pins, copy and short videos for Amazon, Etsy, eBay, Pinterest and TikTok, checked against each marketplace's rules.

    Open Ad Studio

  • Product Studio

    A 15 to 90 second review, demo or promo of your product from your own photos, scene by scene, the product shown exactly as photographed.

    Open Product Studio

  • Listing Studio

    A narrated walkthrough of a home for sale from the agent's photos, with the agent's intro and card.

    Open Listing Studio

  • Editor

    Everything the studios made, and anything you upload, cut together with transitions, titles, narration and music.

    Open Editor
How SceneRender helps

The models make the pieces. The studios make them usable.

A model on its own gives you one result and no second opinion. The platform’s work is everything around that: the plan, the review, the checks, the assembly, and a price before every button.

  • It plans before it makes

    A story becomes scenes, a product becomes a script, a listing becomes a room-by-room tour. You see the plan and change it before anything is generated.

  • You approve every piece

    Scenes, images and takes come back one at a time with their status. Approve what works, remake what does not, and only approved pieces go into the final.

  • It checks the result

    Product shots are compared with your photos; marketplace images are measured against each platform's rules; presenter clips are checked against what was actually said.

  • It keeps the photographed thing honest

    A home or a product is shown as photographed. The camera moves, the setting may change around a product, but the thing itself is not invented.

  • Prices before buttons

    Every generation shows its token cost before it runs, and a generation that fails is not charged.

  • An assistant that can do the steps

    The assistant in the corner answers questions about your project and, when you ask, proposes the next steps with their price; you confirm and it does them.

Technology behind the results

Which model does what.

By what it makes. The names are the ones you may have heard of; the one-line use is what that model does for you here. The exact versions in use are under each group, for those who want them.

Video

Several video models, each chosen for the kind of shot: scenes with recurring characters, ordinary AI shots, scenes with their own sound, presenters who speak to camera, and quick standalone clips.

Seedance (ByteDance)
Scenes with a consistent cast and things you show it photos of: characters, products, equipment, locations. Also Video Studio clips, with or without a reference.
Runway Gen-4.5
Ordinary AI scenes from a description, product shots, and the opening shots of listing videos.
Google Veo
AI scenes that come with their own sound.
Runway GWM-1 avatars
Direct-to-camera presenters: a saved presenter animated to speak your narration, lip-synced.
Kling
Video Studio clips with native audio at three quality levels, and AI edits of a video you already made.
Hailuo
Continuing an approved shot so a story keeps its characters and framing.
Kandinsky
Short five-second clips in Video Studio.
Exact versions in use
  • Seedance 2.0 Fast
  • Seedance 2.0 Standard
  • Seedance 2.5
  • Runway Gen-4.5
  • Google Veo 3.1 and Veo 3.1 Fast
  • Runway GWM-1 Avatars
  • Kling O3 Standard, Kling 3 Standard, Kling 3 Pro
  • Kling O1 Video Edit
  • Hailuo 3.0
  • Kandinsky 5 Pro

Images

Still pictures, product cutouts and the depth that lets a photograph move.

Nano Banana (Google)
Image Studio pictures from a description or from a photo you start with; backgrounds and presenter portraits for ads.
BRIA Product Shot
Placing your product, as photographed, into a generated setting for ads.
BiRefNet
Cutting a product cleanly out of its photo.
Depth Anything
Reading depth from a listing photo so the camera can move through it.
Exact versions in use
  • Nano Banana and Nano Banana Edit
  • BRIA Product Shot
  • BiRefNet V2
  • Depth Anything V2

Writing and planning

The words: scripts, scene plans, listing narration, marketplace copy, captions, lyrics. You edit every line before it is used.

OpenAI GPT models
Product and listing scripts, scene plans and directions, ad copy and captions, song lyrics and style lines in Music Studio.
Claude (Anthropic)
The assistant: answers about the platform and your project, and the steps it proposes and performs on your say-so.
Exact versions in use
  • OpenAI GPT models for writing
  • Anthropic Claude Haiku for the assistant

Voices, music and sound

Narration in stock AI voices, music made to order, and the transcription that checks what a presenter said.

OpenAI speech
Narration in nine stock voices for Clip, Product and Listing Studio; ad voiceovers with delivery instructions. The voices are the platform's; customer voices are not cloned.
OpenAI Whisper
Transcribing presenter clips, word by word, so a clip can be trimmed to the spoken line and checked.
Stable Audio
Music beds under finished videos, plus ambience and sound effects where asked.
Music Studio engines
Four engines at four price points: ACE-Step for quick beats, Google Lyria for clean musical tracks, MiniMax Music for full arrangements with singing, ElevenLabs Music for the most polished sound at an exact length.
Exact versions in use
  • OpenAI tts-1 and gpt-4o-mini-tts
  • OpenAI whisper-1 and gpt-4o-mini-transcribe
  • Stable Audio 2.5
  • ACE-Step
  • Google Lyria 3 Pro
  • MiniMax Music 2.6
  • ElevenLabs Music

Checks and safety

The models that look at results rather than make them.

OpenAI vision
Checks product identity, invented parts, cropping, text and watermarks in ad images; compares product-video frames with your photos; finds readable text in uploaded photos.
OpenAI moderation
Screens every generation request before it runs, so the platform can allow, refuse or hold it for review.
Not AI at all
The Editor's assembly, transitions, titles, audio mixing and the 1440p master are plain video processing (FFmpeg), the same every time.
Exact versions in use
  • OpenAI vision models
  • OpenAI omni-moderation
  • FFmpeg

Models are chosen per job and change as better ones arrive; this page is kept current with the studios. How the platform prompts and combines them is its own work and is not published.

What to expect

Creative tools. Honest expectations.

  • Results vary

    The same request can come out differently twice. That is how these models work; it is why you review each piece and remake what is not right.

  • Some things are refused

    Video models decline scenes that look like a real, identifiable person, and the platform refuses requests that name a real artist or song. A refusal is not charged and does not count against you.

  • Minutes, not seconds

    A scene takes a minute or a few; a music track one to three; a finished video a few more. Your project keeps going if you close the tab.

  • Your media, your rights

    Use photos, videos, likenesses and voices you have the rights to. What you make here is yours to use; the platform labels AI-made media where a platform asks for it.

  • Models change

    When a better model arrives for a job, the studio switches to it and this page changes with it. Your projects and finished videos stay as they are.

Questions & answersMedia rightsCreative guidelines

Start creating.

Sign up, try a studio with the welcome tokens, and see the price before anything runs. Every result on the landing page came out of these same models.

Start free See pricing