2026 Update! This project will be undergoing an overhaul after a release is made for the Linux-CoumfyUI-Launcher repo. There have been significant tech developments in the last year since Studio Whip began that will be bringing major improvements to the entirety of the project.

Studio-Whip

AI-enhanced collaborative content production suite for movies, comics, and interactive visual novels. Create compelling stories with seamlessly integrated tools designed to enhance your creative talents.

There is a mostly incomplete website in construction here, and you can see recent dev proposals/activity in the GitHub issues.

Major Planned Features

• Story-Driven Platform: Create original screenplays with visual story-building tools and advanced LLM integrations.
• P2P Real-Time Collaboration: Create together remotely for free, without requiring third-party servers.
• Storyboarding: Generate images, draw sketches, and position 3D assets to create dynamic animatic storyboards linked to your script.
• Audio Editing: Sequence generated/recorded dialogue, music, and SFX within your timelines.
• Video Production: Develop storyboards into rendered scenes using integrated image and video generation models.
• Professional Color Grading: Make expressive color choices in a color-managed environment using scopes, primary/secondary adjustments, and AI-assisted tools.
• Node-Based Compositing: Combine multiple visual elements (renders, footage, effects) into final shots using a flexible node graph system.

System Requirments

System requirements heavily depend on the size and type of AI models you choose to run locally. You can find many models on platforms like Hugging Face and Civitai.

The table below summarizes recommended hardware specifications for different tiers of usage, focusing on local inference:

• AI Performance (AI TOPS) is measured using FP8 precision.
• CPU performance estimates use PassMark CPU Mark scores.
• Storage estimates are minimums for the Studio-Whip base install, and some models. Your actual needs will be higher depending on the number and size of models and project assets. NVMe SSDs are highly recommended.

Tier	Use Case	RAM	VRAM	AI TOPS	Storage	CPU Performance
Entry-Level	Development, Testing	32GB	8GB	250	32GB	20K+
Mid-Range	Education, Personal Projects	32GB	16GB	500	128GB	30K+
High-End	Advanced Projects, Video	64GB	24GB+	1000	512GB	50k+
Enterprise	Fast Generation	128GB	96GB+	4000	1TB	80K+

Example Model Configurations per Tier

The following table provides example model combinations suitable for each hardware tier when running locally. These are just suggestions; you can:

• Mix and match models based on your specific tasks (writing, image gen, video gen, etc...).
• Use fewer, larger models or more, smaller models depending on VRAM/RAM.
• Choose models optimized for specific hardware (e.g., INT4/FP8 quantizations if supported).
• Combine local models with cloud APIs.
• Distribute models accross CPU and GPU

_{Hover over models for license}

Tier	Creative Writing	Instruct	Image Generation	Video Generation
Entry-Level	Use the Instruct model	Gemma 3 QAT Q4_0 1B BitNet b1.58 2B	Midjourney Mini	Not Practical
Mid-Range	Veiled Calla Q6_K 4B	Gemma 3 QAT Q4_0 4B GLM 4 IQ4_XS 9B	FLUX.1-schnell Q6_K HiDream I1 Fast Q4_K_S Stable Diffusion 3.5 Large Q5_0 SDXL 1.0 Lightning	Wan2.1 T2V 1.3B ltxv-2b-0.9.6-distilled-04-25
High-End	Veiled Calla IQ4_XS 12B MN GRAND Gutenberg Lyra4 IQ4_XS 12B	Gemma3 QAT Q4_0 12B Gemma3 Amoral Q4_K_S 12B GLM Z1 Q8_0 9B	FLUX.1-Schnell Q8_0 HiDream I1 Dev Q5_K_M Stable Diffusion 3.5 Large Q8_0 SDXL 1.0 Base	ltxv-2b-0.9.6-dev-04-25 Wan2.1-12V-14B 480p Q6_K
Enterprise	QWQ ArliAI Q5_K_M 32B QWQ Snowdrop V0 Q5_K_M 32B MN GRAND Gutenberg Lyra4 Q6_K 23.5B Veiled Calla Q8_K 12B	Gemma 3 Abliterated Q6_K 27B Gemma 3 QAT Q4_0 27B GLM Z1 0414 Q5_K_M 32B QWQ Q5_K_M 32B	FLUX.1-Schnell Q8_0 HiDream I1 Full Q8_0	Wan1.2-12V-14B 720p Q8_0

How To Build From Source

Requirements (All Platforms)

• Vulkan SDK: 1.3 or later
• Rust: Latest stable version (via Rustup)
• A GPU Nvidia is sugested due to compatibility and inferance performance, but not strictly required

After installing requirments

1. Clone the repository: git clone https://github.com/<your-repo>/studio-whip.git
2. Navigate to the project: cd studio-whip/rust
3. Build: cargo run --release

For Windows Users

Install Windows Subsystem for Linux, this allows you to run the linux shell script utlities located into /rust/utilities

1. Open Powershell as admin and install wsl wsl --install
2. Find availible linux distros wsl --list --online
3. Install the latest Ubuntu LTS wsl --install --<distro>
4. Launch the Linux distribution: Win+R Ubuntu
5. Windows paths in Ubutu are located in /mnt/<lowercase-drive-letter>/*
6. You may need to install dos2unix within your Linux environment to convert windows line endings.
   • Install it: sudo apt update && sudo apt install dos2unix
   • example usage : dos2unix llm_prompt_tool.sh

After installing the Linux Subsystem, add the Vulkan SDK's glslc compiler to your system variables:

1. Press Win + R, type SystemPropertiesAdvanced, and click Environment Variables.
2. Under "System Variables" or "User Variables," select Path and click Edit.
3. Click New and add: C:\VulkanSDK\<version>\Bin (replace <version> with your installed version).
4. Click OK to save.
5. Verify with glslc --version in PowerShell. It should output the compiler version.

Contributing

Check out the architecture overview, modules documentation, Roadmap, and prompt_tool.sh to get started.

Is This Production Ready?

No. This is a complex early development software with partial and unimplemented features. It will take at least a year to enter plausable production use.