REC Prompt-to-Video Playground
Open Source

Experiment with
AI Video

ComfyDirector is an experimental open-source playground on top of ComfyUI. Give it a prompt, pick a workflow template, keep continuity references close, and try ideas without learning node graphs, wiring models by hand, or treating every output like a finished production.

Cinematic 2.35:1 16:9 Widescreen 9:16 Vertical 4:5 Social
Featured Output

Cyber-Wasteland

Cinematic Lighting 24fps ComfyUI + LTX-2
Format Playground

Pick a frame. Compare the same idea.

Switch between cinematic, widescreen, portrait, and social crops to see how framing changes the result. It is a quick way to test composition before you commit to a workflow.

sample clips

Ratio
Focus
Feel
Garden Atmosphere 2.35:1
More room for atmosphere
Use this when the environment should carry real weight instead of just sitting behind the subject.
Featured Output 2.35:1
City and subject can coexist
The wide crop keeps the skyline readable without shrinking the action into the distance.
Character Scene 2.35:1
Good for slower scenes
Extra width helps a shot breathe when pacing matters more than speed.
Puppy Clip 16:9
Balanced and easy to read
16:9 is the easiest place to test an idea when you want subject, motion, and background to all stay clear.
City Experiment 16:9
A strong default frame
It is usually the safest starting point when you want quick comparisons without dramatic cropping.
Scene Texture 16:9
General-purpose widescreen
If you are not sure which frame to try first, this one usually gives the cleanest baseline.
Studio Portrait 9:16
Portrait puts the subject first
The tall crop cuts side noise and keeps the subject dominant.
Anime Alley 9:16
Vertical pushes focus forward
It reads fast on phones and makes faces or figures carry more of the shot.
City Crop Study 9:16
A fast reframing test
Useful when you want to see how a wide source behaves as a phone-first vertical.
Social Puppy 4:5
Compact without going full vertical
4:5 keeps the main subject large but still leaves some breathing room around it.
Studio Feed Cut 4:5
Good for feed-style previews
Useful when you want a social-friendly crop without committing all the way to 9:16.
Color Study 4:5
A practical social middle ground
It trims the edges, keeps the center strong, and lands between widescreen and portrait.
About

Why I Built This

ComfyDirector started as a way to mess with AI video without constantly wrestling with raw node graphs.

It wraps ComfyUI in a simpler experimental layer for storyboard review, template routing, continuity references, refinement, audio, and assembly. The point is not to pretend this is polished studio software. The point is to make it easier to play, compare, rerun, and learn what works.

Workflow Experiments

A lighter way to mess around with prompts, templates, references, refinement, audio, and workflows without learning raw ComfyUI first.

Highlighted Feature

Continuity is now part of the workflow.

The Library lets you save characters, props, locations, styles, and motion notes, link them into projects, and bind them to scenes. Supported templates can route a primary character image directly into image generation.

Library
Reusable references
Bindings
Project, scene, angle
Visual Input
Character image routing
Review
Usage provenance

LLM Prompt Expansion

Start with a rough idea and let the built-in LLM turn it into scenes, shot directions, and prompts you can actually test.

Storyboard Review

See scene beats, images, reference usage, and dialogue choices before you commit to heavier video runs.

Continuity Library

Promote useful images into reusable references, curate them globally, and bind characters or other continuity notes back into scenes.

Refine Studio

Use opt-in image refinement, manual masks, and assist flows to repair or redirect shots before sending them into video generation.

Format Playground

Swap between cinematic, widescreen, portrait, and social framing to see how the same idea changes.

Execution Trace

Prompts, images, videos, audio, workflow exports, reference metadata, and rerun history stay attached to each project.

Built-in Editor

Make a rough cut inside the app. Reorder clips, trim them, add transitions, adjust SFX mix, and queue optional movie mastering.

How the Workflow Works

1

Prompt

Write the idea in plain English. No node graph needed.

2

Expand

AI turns it into scenes you can review, tweak, and approve before the heavy stuff runs.

3

Generate

Pick a template and let ComfyDirector run the matching ComfyUI workflow behind the scenes.

4

Assemble

Clips and audio can be stitched into a simple final video.

Built for messing around: rerun stages, swap templates, tweak prompts, change seeds, and keep going until something interesting shows up.

Available Workflows

11 Gen 2 templates across 5 GPU tiers — from lighter 8GB experiments to external/self-hosted LTX 2.3 cinematic, I2V, and lip-sync workflows.

8 GB RTX 4060, 3070

Quick Video

1 step

Fast text-to-video with Wan 2.2 5B. Sound design via MMAudio. Perfect for quick experiments on modest GPUs.

Prompt Video Sound
Sound Design
16 GB RTX 4080, A4000

Audio+Video

3 steps

Synchronized audio and video in a single pass with LTX-2. Camera control, face enhancement, and LoRA support built in.

Image Camera Video+Audio
Native Audio Camera LoRA Face Detailer
20 GB RTX 4070 Ti Super, A5000

Studio Video

1 step

Dual-expert MoE architecture delivers the highest quality text-to-video. LoRA support for style customization.

Prompt Video
LoRA Sound Design

Anime Studio

1 step

Specialized anime pipeline with curated LoRAs: retro 90s, modern HD, and Kawajiri styles. Built on the 14B MoE backbone.

Prompt Anime Video
LoRA Sound Design
24 GB RTX 4090, A5000 Ada

Cinematic Film

5 steps

The complete cinematic pipeline. Hybrid video (S2V lip-sync + LTX-2 motion), TTS narration, multi-character support, face enhancement, and opt-in image refinement before video.

Character Image Camera TTS Video
Hybrid Video TTS Camera LoRA Face Detailer Refinement Sound Design
32 GB+ External self-hosted preferred

Audio+Video 2.3

3 steps

Experimental LTX 2.3 native audio+video. Keeps the Qwen image and camera stages, then upgrades the final pass to newer synchronized dialogue and ambient audio generation.

Image Camera Video+Audio
Native Audio Dialogue Camera LoRA Experimental

LTX Cinematic

1 step

Direct LTX 2.3 text-to-video with synchronized audio. Best when you want the newest cinematic backend without the full multi-stage image and camera pipeline.

Prompt Video+Audio
Native Audio LoRA Refinement Experimental

LTX Cinematic I2V

2 steps

Single-image LTX 2.3 image-to-video with storyboard image upload fallback, native synchronized audio, and two-pass cinematic refinement.

Image Video+Audio
I2V Native Audio Refinement External Preferred

LTX I2V Sequenced

2 steps

Guided LTX 2.3 image-to-video with up to five visual anchors per clip through the WhatDreamsCost sequencer workflow pattern.

Anchors Sequencer Video
Multi Image Guided Motion Experimental External Preferred

LTX Cinematic Lip-Sync

3 steps

AITold-style cinematic lip-sync from one scene image and uploaded or generated dialogue audio, with guide-audio separation and two-pass refinement.

Image TTS/Audio Lip-Sync
Audio Upload TTS Cinematic Refinement External Preferred

LTX TalkVid Lip-Sync

3 steps

Routed LTX 2.3 ID-LoRA talking-head clips from storyboard images plus uploaded or generated dialogue, with Qwen or ZImage fallback.

Image TTS/Audio TalkVid
Audio Upload TTS ID-LoRA LoRA Experimental

Plus 6 legacy templates (Gen 1) still available for backward compatibility.

System Requirements

ComfyDirector runs locally if you have enough GPU headroom. The lighter templates start around 8GB VRAM. The newest LTX 2.3 I2V and lip-sync routes are best treated as external/self-hosted 32GB+ workflows today, while the managed Docker/WSL runtime remains below recommendation for the heaviest LTX 2.3 paths. Most hands-on validation has been on an RTX 5090.

If your machine is smaller, start with the lower-VRAM templates or use cloud GPUs for heavier runs. The managed ComfyUI service runs through Docker on Linux or WSL2; if you already have a native Windows or Linux ComfyUI install, point ComfyDirector at it as an external runtime.

Read detailed Hardware Guide
GPU VRAM 8GB+ (32GB external / 48GB managed for heavy LTX 2.3)
System RAM 32GB+ (64GB ideal for LTX 2.3)
Disk Space Variable (Depends on installed templates)
Platform Managed: Linux/WSL2 (external ComfyUI: Windows/Linux)

Support the Project

ComfyDirector is an open-source side project built in spare time. If it saves you from wrestling with raw ComfyUI just to test a prompt, Buy Me a Coffee helps cover the time and experiments behind it.

If you would rather not donate, starring the repo, opening issues, and contributing workflows or templates are still extremely useful signals.