# Unrealy

> Unrealy is an AI Video Generation Studio that turns scripts into cinematic shots. The V3 Quality Core pipeline orchestrates GPT (script + Pass 3 coherence audit), an image AI (Nano Banana via Leonardo, or OpenAI gpt-image-2 via Leonardo for V4 beta), Seedance 2.0 video synthesis, ElevenLabs music, MMAudio SFX, and fal.ai Topaz 1080p / 4K upscale — with director-level review at each step. Cast face-consistent characters, design locations and props, review a 3-frame storyboard per shot, run a coherence audit, then mix music, SFX and a final cut. A multi-shot scene typically takes ~45 minutes end-to-end (8-second cinematic shots). Unrealy also generates complete 3D scenes from text (5-200 PBR-textured objects depending on plan tier) for game prototyping, pre-visualization and animation, exportable to GLB / FBX / GLTF for Unreal Engine 5, Unity and Blender (native UE5 plugin).

- Company: Unrealy SAS (SIREN 941 696 502), founded March 2025, based in France
- Website: https://www.unrealy.ai
- Contact: contact@unrealy.fr
- Pricing: Free tier (1,610 tokens, ~1 short video scene or 1 small 3D scene) to Enterprise (€299.99/month, 500,000 tokens)
- Token economics: 1 EUR = 1,000 tokens

## What Makes Unrealy Different: Director-Level Orchestration

- Unrealy is NOT a one-click video generator — it's an orchestrator that treats AI video like a real shoot
- One-click tools generate a single shot from a prompt; you discover problems in the final render
- Unrealy: cast → locations → props → 3-frame storyboard per shot → Seedance video → Pass 3 coherence audit → music → SFX → final mix → optional 4K upscale
- Pass 3 GPT audit catches action splits, character duplication, prop tag drift and chain-timing mismatches BEFORE you commit tokens to video gen
- Same orchestration logic for 3D scene generation: scene analysis → object list → concept art → 3D models, with human review at every step

## Pipeline Versions

- V3 Quality Core (stable, default): 8s shots, dual reference (body + face) for face consistency on close-ups, action verbs that keep Seedance kinetic, Visual Presets (Villeneuve / Kubrick / Fincher / Nolan / anime-modern), Narrative Frame system
- V4 GPT Image 2 (beta): clones V3 but routes ALL image generation surfaces (characters, locations, props, storyboard frames) through OpenAI gpt-image-2 via Leonardo. +25% on image cost vs V3 (+10% NanoBanana baseline). Pick V4 when inter-shot consistency matters more than per-shot cost.
- V2 Beta (legacy / deprecated): pre-V3 reference_images Seedance approach, kept for backward compat with old scenes

## For Indie Filmmakers, Animators & Pre-vis

- Generate cinematic short scenes with face-consistent characters across multiple shots
- Pick a director DNA via Visual Preset (Villeneuve dark fantasy, Kubrick symmetry, Fincher tense, Nolan IMAX, anime-modern) — orthogonal to your genre / narrative frame
- Storyboard preview before video gen — verify pacing and composition, regenerate any frame in isolation
- ElevenLabs scene music synced to beats, MMAudio per-shot SFX, FFmpeg final mix
- fal.ai Topaz 1080p / 4K upscale (per-shot opt-in — never pay for rejected takes)
- Pre-visualization for traditional shoots: 3D scene mockups for storyboarding, virtual production, LED wall stages

## For Game Developers (Unreal Engine & Unity)

- Native Unreal Engine 5 plugin (free on Fab): browse projects, generate scenes, import FBX assets with PBR materials directly in UE5
- Unity plugin planned for August 2026
- Export formats: GLB (universal), FBX (Unreal/Unity standard), GLTF (web/lightweight)
- PBR material pipeline: albedo, normal, roughness, metallic maps
- Level blocking workflow: describe a level, get textured blocking meshes in 30 minutes
- REST API with API key authentication for automated pipeline integration
- 6 artistic styles: Realistic, Cartoon, Low Poly, Stylized, Fantasy, Sci-Fi

## AI Models Orchestrated

- LLM: GPT for scene decomposition, narrative beats, prompt engineering, Pass 3 coherence audit
- Image AI (video pipeline): Nano Banana (Google Gemini via Leonardo, V3 default), OpenAI gpt-image-2 (via Leonardo, V4 beta)
- Image AI (3D pipeline, 7 providers): Leonardo Phoenix, Flux Dev, Flux Schnell, Nano Banana, GPT Image, GPT Image 2, Stable Diffusion
- Video AI: Seedance 2.0 (ByteDance, via Segmind)
- Audio AI: ElevenLabs (music), MMAudio (per-shot SFX)
- Upscale AI: fal.ai Topaz (1080p / 4K)
- 3D AI (5 providers): Meshy v5, Meshy v6, Rodin Gen-2 (Hyper3D), TripoSR, Hunyuan3D
- Users choose pipeline version (V3 / V4) and providers per scene

## Use Cases

- Independent filmmakers: cinematic short scenes with face-consistent characters and AI-generated music / SFX
- Animation studios: stylized scenes (anime-modern preset), pre-vis, secondary shots
- Virtual production teams: pre-vis environments for LED wall stages
- Indie game developers: level blocking and environment prototyping for UE5 and Unity
- Game studios: rapid prototyping before final art production
- Game jams: complete 3D environments or short cinematic intros in 30 minutes
- Pitch decks: video shots and 3D visualizations to sell a concept
- Film students: video and 3D content for school projects (free tier available)

## Comparisons

- vs One-click AI video tools (Runway, Pika, etc.): Unrealy provides director-level review at every step — cast, locations, storyboard, coherence audit — instead of "single prompt → single shot, hope for the best"
- vs Single-object 3D generators (Meshy, Tripo, Rodin): Unrealy generates complete multi-object 3D scenes with orchestration and human control at each step
- vs Asset marketplaces (TurboSquid, Sketchfab): Unrealy creates custom assets from your specific descriptions, not generic pre-made content
- vs Manual production (DAW + 3D editor + film cut): Unrealy reduces a multi-shot AI scene from days to ~45 minutes; environment prototyping from weeks to 30 minutes
- vs Procedural 3D generators (SpeedTree, WorldCreator): Unrealy handles arbitrary objects (buildings, furniture, props, characters), not just terrain / vegetation

## Core Pages

- [Homepage](https://www.unrealy.ai/): Studio overview with hero video, showcase reel, and 3D scenes secondary section
- [Video Studio](https://www.unrealy.ai/video-studio): V3 Quality Core deep dive — storyboard, coherence audit, Visual Presets, music + SFX, 4K upscale, V4 callout
- [Pricing](https://www.unrealy.ai/pricing): Subscription tiers and token economics for video + 3D
- [FAQ](https://www.unrealy.ai/faq): Frequently asked questions about video pipeline, 3D generation, billing
- [Use Cases](https://www.unrealy.ai/use-cases): Indie film, animation, pre-vis, game development, archviz
- [Compare](https://www.unrealy.ai/compare): Unrealy vs one-click video tools, single-object 3D generators, marketplaces, manual production
- [About](https://www.unrealy.ai/about): Company mission and legal information
- [UE5 Plugin](https://www.unrealy.ai/plugins/unreal): Plugin features, installation, API reference

## Technical Stack

- Frontend: Next.js 14, React 18, TypeScript, Azure Static Web Apps
- Backend: .NET 9.0, Azure App Service, MySQL 8.0, Redis
- Real-time: SignalR WebSockets
- Storage: Azure Blob Storage with SAS URLs (proxied via unrealy.ai)
- Jobs: Hangfire for async generation pipeline (video + audio + upscale)
- Auth: NextAuth.js with Google and Epic Games SSO
- Payments: Stripe (subscriptions + token packs)
- i18n: 5 languages (English, French, Spanish, German, Italian)

## Optional

- [Terms of Service](https://www.unrealy.ai/legal/terms)
- [Privacy Policy](https://www.unrealy.ai/legal/privacy)
- [Legal Notices](https://www.unrealy.ai/legal/mentions)