Original AI studio film · 45 seconds · Three connected LTX takes · Silent. Play the journey, or explore the studio below.
MAKE THE
IMPOSSIBLE.
Human-directed video ads and short films,
made with AI.
YOU HAVE A STORY.
LET’S MAKE IT MOVE.
For a product to remember.
Or a world to get lost in.
VIDEO ADS
Turn your product, launch, or brand message into a film with a point of view. Built around what your audience needs to notice, feel, and do.
Campaign films · Product launches · Social cutsSHORT FILMS
A character. A conflict. A world worth staying in. We develop the story and direct the images, performance, and edit around it.
Narrative concepts · Storyboards · Film productionEvery production is scoped and quoted for your brief.
A SEAT FOR
ELSEWHERE.
An ordinary chair. An extraordinary feeling.
One connected journey. A story with a point.
45-second studio journey · Silent · Scroll to explore
AI can create the world.
Direction gives it a point.
POSSIBILITY NEEDS
A POINT OF VIEW.
That’s why you work with a studio. We start with the story, direct the images around it, and make the decisions that give a film its feeling.
The idea leads. The technology follows.
See how we workHUMANS AT
EVERY CUT.
More possibility.
Still a deliberate process.
- 01
Brief
You tell us the story you need, the audience and the budget. We listen before we pitch.
- 02
Concept and story mode
Your idea becomes a logline, beats, tone and visual direction, agreed before a frame exists.
- 03
Board
Key frames and shot order, laid out for your sign-off. The film is decided here, not in the edit.
- 04
Shoot and generate
Our in-house environment drives the leading video, image and audio models, a director on every shot.
- 05
Edit and grade
Editors cut for rhythm and performance, then grade and mix until it plays as one film.
- 06
Delivery
The finished film, mastered and handed over in the formats your channels need.
THE TOOLS CHANGE.
THE TASTE IS OURS.
Our in-house production environment brings video, image, and audio engines together. We choose the tool to suit the shot.
Explore the engines we work with
Many models. One in-house studio.
We built our own in-house production environment, where our storytellers direct the major video, image and audio models side by side, from Veo, Kling and Seedance to LTX, Wan and Runway. The right model is chosen shot by shot; the story stays human.
11 of the video models are Chinese; we drive them beside the Western ones.
Video · West
13- Veo 3.1US · APIGoogleGoogle DeepMind video model with native audio; in Gemini API, Vertex AI and Flow.
- Runway Gen-4.5US · APPRunwayText and image to video from Runway.
- Runway AlephUS · APPRunwayIn-video editing model: change angle, lighting or objects in existing footage.
- LTX-2IL · OPENLightricksAudio-and-video model with public weights.
- Luma Ray3US · APPLuma AIDream Machine's video models; Ray3 adds HDR.
- PikaUS · APPPikaVideo generator with Scene Ingredients for your own characters and objects.
- MochiUS · OPENGenmoOpen-weights text-to-video model from Genmo.
- Midjourney VideoUS · APPMidjourneyImage-to-video from your Midjourney stills.
- MareyUS · APPMoonvalleyVideo model Moonvalley says is trained on licensed footage; camera control.
- Grok ImagineUS · APIxAIxAI's image and video generator; video clips come with audio.
- Firefly VideoUS · APPAdobeAdobe's video model, inside Firefly and Premiere Pro alongside partner models.
- HedraUS · APPHedraTurns a still image plus audio into a speaking, performing character video.
- Stable Video DiffusionUK · OPENStability AIImage-to-video model with public weights; an earlier open baseline.
Video · China
11- Kling 3.0CN · APIKuaishouMulti-shot video with native audio.
- Seedance 2.0CN · APIByteDanceVideo model driven by text, image, video and audio references.
- Hailuo 3CN · APIMiniMaxMiniMax H3 video model with native audio.
- WanCN · OPENAlibabaOpen-weights video model family from Alibaba.
- HappyHorseCN · APIAlibabaAlibaba video model: text, image and reference to video.
- HunyuanVideoCN · OPENTencentOpen-weights video family with a variant sized for consumer GPUs.
- ViduCN · APIShengshuVideo line with reference-to-video and synced audio.
- PixVerseCN · APIPixVerseMakes multi-shot clips with synchronized audio.
- CogVideoXCN · OPENZhipu AIOpen-weights text-to-video family from Zhipu AI.
- SkyReelsCN · OPENSkyworkSkyReels V3 open weights; reference, audio and video inputs.
- OmniHumanCN · APIByteDanceAnimates a single photo of a person from audio or pose input.
Image
11- MidjourneyUS · APPMidjourneyImage generator on web and Discord; the V8 series is the current default.
- FLUX.2EU · OPENBlack Forest LabsImage generation and editing family: open 32B dev weights, hosted Pro and Flex.
- GPT Image 2US · APIOpenAIOpenAI image model; in ChatGPT and the API as gpt-image-2.
- Nano BananaUS · APIGoogleGoogle's Gemini image models, Nano Banana 2 and Pro; generation and editing.
- Ideogram— · APPIdeogramImage generator with a focus on legible text inside images; web app and API.
- SeedreamCN · APIByteDanceByteDance image model line, available through its API.
- Qwen-ImageCN · OPENAlibabaImage generation and editing model line from Alibaba's Qwen team.
- HunyuanImageCN · OPENTencentOpen-weights image model from Tencent's Hunyuan team.
- RecraftUS · APPRecraftImage generator for brand and design work, including vector (SVG) output.
- Stable DiffusionUK · OPENStability AIOpen-weights image family that runs locally and supports fine-tuning.
- ReveUS · APPReveImage generation and editing model.
Audio
06- ElevenLabsUS · APIElevenLabsVoice and music generation; Eleven v3 handles expressive speech in 70+ languages.
- SunoUS · APPSunoSong generation with vocals from a text prompt; v5 and later models.
- UdioUS · APPUdioSong generation platform for text-to-song and remixing.
- LyriaUS · APIGoogleGoogle's music model; Lyria 3 Pro makes tracks up to three minutes.
- Stable AudioUK · OPENStability AIStable Audio 3: music and sound-effect models, three with open weights.
- MiniMax SpeechCN · APIMiniMaxText-to-speech in HD and Turbo variants, with voice cloning.
Studios & tools
17- HiggsfieldUS · APPHiggsfieldCreator platform with cinematic camera controls over several video models.
- FreepikEU · APPFreepikDesign platform with image and video generators, including partner models.
- KreaUS · APPKreaReal-time image and video generation with editing and upscaling tools.
- LTX StudioIL · APPLightricksStory-to-storyboard-to-video workspace built on Lightricks' LTX models.
- Leonardo— · APPLeonardo.AiImage and video generation platform with fine-tuned model options.
- HeyGenUS · APPHeyGenAI avatar and video translation platform with lip-synced speech.
- SynthesiaUK · APPSynthesiaAI presenter video platform used for training and corporate video.
- TopazUS · APPTopaz LabsUpscaling, denoise and frame-interpolation tools for images and video.
- MagnificEU · APPMagnificAI image upscaler and detail enhancer.
- ComfyUIUS · APPComfy OrgOpen-source node-based workflow engine for running media models.
- fal.aiUS · APIfalHosted inference API for image, video and audio models.
- ReplicateUS · APIReplicateHosted API for running open and proprietary media models.
- Kling AICN · APPKuaishouKuaishou's web and mobile app for Kling video and image generation.
- Hailuo AICN · APPMiniMaxMiniMax's web and mobile app for Hailuo video generation.
- DreaminaCN · APPByteDanceByteDance's creative app for Seedance and Seedream generation.
- RunwayUS · APPRunwayRunway's web app and API for generation, editing and Aleph.
- FlowUS · APPGoogleGoogle's filmmaking workspace for Veo-based scene building.
Names and versions are a reference list of models we work with, not endorsements or partnerships. All trademarks belong to their owners.
EVERY FILM STARTS
WITH A SCRIPT.
The next scene is yours.
Tell us the idea, the audience, and the feeling.
We start with the concept and script.
Two speeds. One studio.
The full film
Tell us the film you need, the budget and the deadline. The team reads it and contacts you to scope and quote the production.
Send a briefQuoted per project.
The Concept plan
$399One payment · 3 ideas in · 3 concepts out
One payment to test the studio before you commit. Bring three ideas, get three concepts in story mode with two ideations each, then commission the full film if one of them is the one.
- 01Three concepts, one for each idea you share
- 02Two ideations for every concept
- 03Story mode for each concept: logline, narrative beats, tone and visual direction
- 04Full-film production is optional and quoted per project
Send a brief with a budget.
Tell us the film you need, the budget and the deadline. The team reads it and contacts you to scope and quote the production.
- 01 · WE READ IT
Every brief is read by a person on the team, start to finish, before anyone replies.
- 02 · WE CALL YOU
We talk through the film, the budget and the deadline until the scope is clear to both sides.
- 03 · WE QUOTE THE FILM
You get a quote for the production, scoped to your project and not to a rate card.
Or write to illusionartstudios@gmail.com