The library
ComfyUI workflows.
35 workflows
MiniMax H3newCharacter Video Swap — one photo, any performance
A guide video gives the moves. Your photo gives the person.
MiniMax H3newH3 Image Edit — up to 8 references
Edit a picture with other pictures, in 8 steps.
Krea 2newBulk Character Swap — Dynamic LoRA
A folder of photos in. Your character in every one — and the right LoRAs picked per picture.
Krea 2newBulk Reference Swap — no training
One reference photo. Every picture in a folder, re-cast — no LoRA needed.
Krea 2Bulk Character Swap
The straight version: a folder in, your character out, one knob.
LTX 2.5Long Video Upscale — LTX 2.5
The whole clip, not the first three seconds. 2× and sharper, audio kept.
MiniMax H3Character Portrait — one photo
A new photo of a person from a single reference picture. No training.
MiniMax H3newFlow Long — 10 to 30 seconds in one job
One job, one file, continuous sound.
MiniMax H3newH3 Image — P-H3, 8 steps
H3's eye for light and skin as a still-image model.
MOSS-ImagenewMOSS-Image
The open model that topped Civitai this month. Apache 2.0.
MOSS-ImagenewMOSS-Image Layout — posters, covers, title cards
Draw the boxes. MOSS puts the words and the people where you said.
Krea 2newKrea 2 IntoRealism Flash — 4 steps
IntoRealism in 4 steps.
Krea 2newKrea 2 CyberRealistic — 8 steps
Photographs, not renders.
ACE-StepnewSong from an idea — ACE-Step
A second song engine. Apache 2.0, 8 steps.
VoicenewVoice Clone — speak in your voice
Ten seconds of your voice. Any text, 23 languages.
VoicenewVoice Clone — any song in your voice
Same song, same backing track, your voice singing it.
VoicenewVocal Split — a voice sample from any song
Pull the lead vocal out of a finished song.
Krea 2newRandom Influencer Face Generator
A new influencer face every time you press Queue: the first step to your own character.
Krea 2newKrea 2v2
Our Krea 2, upgraded: a realism LoRA plus a detail pass, 1536×1920 out.
Krea 2Krea 2 IntoRealism
The community realism fine-tune, without the confetti and speckles.
Z-ImageZ-Image Turbo
Fast photoreal with face and eye detailing and a SeedVR2 finish.
FLUX.2FLUX.2 Multiple Characters
Four reference photos in, one group shot out — locally, no API.
MiniMax H3Character Ensemble
Several people from several photos, in one picture.
MiniMax H3Character Ensemble Video
The same cast, moving — with sound.
MiniMax H3Flow — text, image or first/last frame to video
One video template, three modes.
MiniMax H3Flow Continuation — longer shots
Chain clips into one continuous take.
MiniMax H3Flow Extend — seamless longer takes
Extend a take without a visible join.
MiniMax H3Flow Image — stills from a video model
A photoreal still out of MiniMax H3.
LTX 2.5Video Upscale — LTX 2.5
2× latent upscale and refine for a short clip.
UtilityVideo Upscale — ESRGAN
Sharpen any clip the cheap way, any length.
UtilityFrame Interpolation
Smoother motion from in-between frames.
SeedVR2Image Upscale — SeedVR2
Detail-restoring upscale for stills.
MiniMax MusicSong from an idea
Idea in, finished song out.
MiniMax H3Music Video
A song and a recurring character, cut into a real video.
UtilityRegister a Character
One record per character, used by every workflow.