figaRRRo Studio

The figaRRRo difference

Your screenplay drives everything

Most AI film tools start with a blank prompt. figaRRRo starts with your script — and builds everything else from it automatically.

📜

Import your screenplay

Drop a .fdx, .fountain, .pdf or .txt file. figaRRRo parses every scene, character, line of dialogue and action description — in seconds.

↓

🎭

Characters, locations, wardrobe & props — extracted automatically

Every character that appears in the script is added to your cast. Scenes, wardrobe items and props are scanned and linked to the right characters and shots.

↓

✦

Biographies & visual profiles — one click

figaRRRo reads the screenplay and auto-generates detailed character biographies — personality, backstory, physical description, outfit — ready to use as AI prompts or hand off to a casting director.

↓

🖼️

Reference images — import or generate

Upload photos of your characters, or generate them directly inside figaRRRo — the same way as the Chrome extension. Character, location, wardrobe and prop references are all stored against each shot, not scattered across folders.

↓

🎯

Shot prompts with all references ready

figaRRRo analyzes each scene and writes First Frame + Last Frame prompts — with character references, location, wardrobe and props already attached. No copy-pasting between tabs.

↓

🎬

Generate inside figaRRRo — or take the prompt anywhere

Render frames and videos directly via built-in providers (Atlas Cloud, fal.ai, Google AI, OpenAI, ComfyUI), or copy the prompt and generate in Midjourney, Runway or any tool you already use. Either way, everything lands back in your project — organized, named, exported.

See it in action

Real project. Real output.

From screenplay import to cinematic frame — everything organized in one place.

figaRRRo Studio — generated cinematic frame, Tuscan coast road

AI-generated frame from the shot editor — characters, location and wardrobe references all attached

✦ Character bio auto-generated from screenplay

Shot editor with frame prompt, scene description and references

🎯 Shot editor — scene prompt with all references ready

Wardrobe item with reference image, assigned to characters

👗 Wardrobe items linked to characters & scenes

Props management — vehicle with reference image, linked to scene

🚗 Props & vehicles with reference images, per scene

How it works

Three ways to work — pick yours, or mix

No lock-in. Switch between modes per shot, per scene, or per project.

① All-in-one

Everything inside figaRRRo

Drop a screenplay, generate frames + clips + voices via API providers (Atlas Cloud, Google AI, fal.ai, OpenAI, Anthropic, ElevenLabs) or use your own GPU via ComfyUI.

② Hybrid hub

figaRRRo manages, you generate

figaRRRo handles your script, characters, shots, references and prompts. Generate frames in Midjourney, Runway or Weavy. Drop renders back by filename — auto-matched.

③ Local / open-source

Bring your own GPU

Flux, SDXL, Wan 2.5, LTX 2.3, Z-Image Turbo, custom LoRAs via ComfyUI. figaRRRo submits workflows with reference images injected and pulls results back into the project.

What it does

End-to-end production, in one window

Every tool you need to take a screenplay from first draft to final export.

📜

Screenplay Intelligence

Import .fdx, .pdf, .fountain or .txt. Auto-extract scenes, characters, dialogue, action lines. Scan for wardrobe and props with one click.

🎭

Character Library

Reference photos with AI-analyzed visual profiles — gender, ethnicity, hair, outfit, signature items. Library + Bio + Generate + Voice + Wardrobe tabs per character.

🎨

Frame Generation

Nano Banana 2 (Google Gemini 3.1 Flash Image), GPT Image 2 (OpenAI / fal.ai), Flux 1.1 Pro, Seedream, Reve — or your own GPU via ComfyUI. Multi-ref: character + wardrobe + location in one shot.

✎

Storyboard Sketches

B&W pencil sketches via Z-Image Turbo (Tongyi · 8-step) at ~$0.005 each, or free via ComfyUI. Review composition before committing to full-res frames.

🎞

Multi-Shot Timeline

Up to 9 keyframes per shot — Seedance 2.0 reference-to-video or Kling O3 Multi-Shot. Suggest Beats lets AI split your scene into 2–6 timed beats automatically.

🎬

Video Animation

Kling 3.0 V3 Pro, Kling O3 Multi-Shot, Seedance 2.0 (cloud) or Wan 2.5 / LTX 2.3 (local ComfyUI). Live cost preview before every render.

🎛

Beat Prompts & @-Tokens

Write shot prompts with colour-coded @-tokens — @Character, @Location, @Wardrobe inline pills. See exactly which reference each beat hits, before you render.

🔊

Audio Pool

Voice clips, SFX and BGM per shot — all in one pool. Generate voice inline with Gemini TTS (free tier · 30 voices), ElevenLabs or MiniMax. Attach clips to beats, ship to Seedance or Kling.

🎤

Lip-Sync

Happy Horse 1.0 (Alibaba · top leaderboard), LatentSync, Sync 2.0 Standard / Pro for hero close-ups. Auto-suggests the matching voice clip for each shot.

🎬

Director's View

Three tabs per shot — Storyboard / Frame / Video — so sketch, still and clip live in one place. Camera Control lets you shift angle, then re-render with one click.

⚙️

3-Step Settings Setup

Tick your providers → paste keys inline → assign models per task. Atlas Cloud + Google AI = cheapest viable combo — two keys cover image, video, LLM and TTS.

🎯

Visual Style Guide

Upload a director's treatment PDF or paste a style guide. LLM extracts color palette, lighting, camera character — and auto-applies to every prompt downstream.

🔄

Round-trip with any tool

Copy-prompt buttons on every screen. Filename convention (1a_first.jpg) so renders from Midjourney, Runway or Weavy auto-match the right shot when dropped back in.

📦

Export

One-file HTML Production Bible (base64 images), PDF storyboard, Markdown / JSON / CSV shot lists. Drop into Notion, hand off to a DOP, or feed your own pipeline.

Real workflows

Two ways from screenplay to export

Both produce the same project file. Mix them per shot.

Workflow A

End-to-end inside figaRRRo

1Drop .fdx — scenes, characters, dialogue extracted in seconds

2Upload 2–3 reference photos; ✦ auto-fills the visual profile

3✦ Analyze Scene writes First / Middle / Last prompts

4✎ Storyboard ($0.005) → ✦ Frame ($0.07) → 🎬 Video (Kling O3 Standard $0.56)

5🔊 Generate All — every dialogue line through your TTS provider

6🎤 Lip-Sync — LatentSync default at $0.20 aligns mouth to voice

7📦 Export HTML Bible + PDF storyboard

Workflow B

figaRRRo as hub, generate outside

1Same screenplay import + character library setup

2✦ Analyze Scene writes differentiated prompts as before

3📋 Copy full Frame prompt → paste into Midjourney / Runway / Krea

4Save with figaRRRo filename convention (e.g. 1a_first.jpg)

5Drop onto Story → Import Frames — auto-matched to the right shot

6Repeat for video: copy prompt → Runway / Kling → drop .mp4 back in

7📦 Export HTML Bible + PDF storyboard

Pricing

Free during beta. You pay only for APIs.

figaRRRo Studio is free. You bring your own API keys and pay the providers you choose, directly, at their current rates. Nothing flows through figaRRRo — no markup, no middleman.

Atlas Cloud ⭐

Image · Video · LLM — one API, cheapest-per-tier across the board. Nano Banana 2, Z-Image Turbo, Seedance 2.0, DeepSeek V4 Flash.

Google AI

Nano Banana 2 frames · Gemini 2.5 Flash LLM · Gemini TTS (30 voices, free tier up to 1M chars/month).

fal.ai

GPT Image 2 · Z-Image Turbo storyboards · LatentSync · Happy Horse 1.0 lip-sync · Flux · Seedream.

OpenAI

GPT Image 2.0 — multi-style image generation, directly via OpenAI API or fal.ai.

Anthropic

Claude — LLM for scene analysis, biography generation and beat splitting.

ElevenLabs

Premium voice synthesis + voice cloning per character.