Fuobe is built on open models, open code and the people who share them. Thank you. Here is each one, who made it, what Fuobe uses it for and its licence.
Thank you, Comfy Org
ComfyUI
Every open model in Fuobe runs through ComfyUI, on your own GPU. It is free and open, and Fuobe would not exist without it.
Hosts the open model files Fuobe runs
The labs and people who publish the open models Fuobe runs. Thank you for sharing them. Tap one to see where Fuobe uses its models.
Some licences ask us to show their name, like a “Built with” line. We're glad to.
MiniMax H3
Built with Llama
In HiDream-I1 Dev
MiniMax-Music3
In MiniMax Music 3
Built with DINOv3
Powered by Hi3D 1.0
In Hi3D 1.0
MiniMax H3 is licensed under the MiniMax H3 Community License Agreement, Copyright © 2026 MiniMax. All Rights Reserved. FL2VA and Ref2VA adapters Copyright © 2026 Mellis, Inc. These adapters are Model Derivatives of MiniMax H3 and are distributed under the MiniMax H3 Community License Agreement. Files modified from MiniMax H3 carry notices stating that they have been modified.
Fuobe Soul
Fuobe Original
139 signature looks, ready to shoot
Fuobe Soul HD
Fuobe Original
Fuobe Soul's looks with clean lettering for posters and signs
Fuobe Soul Cinema
Fuobe Original
One film look
Fuobe Marketing Image
Fuobe Original
Product shots, ads, posters and listings from your product photo
Qwen Research License(opens in a new tab)
Non-commercial use only
Fuobe Try-On
Fuobe Original
Puts the garment in image 2 (a bottom in image 3) on the person in image 1
Fuobe Angles
Fuobe Original
The same shot from another camera
Auto
Fuobe
Picks the model for you
Krea 2 Community License(opens in a new tab)Qwen Research License(opens in a new tab)
Non-commercial use only
Ideogram 4
Ideogram
Typography, posters and photographic stills from structured captions
Ideogram Non-Commercial Model Agreement(opens in a new tab)
Non-commercial use only
Qwen-Image 2.1
Alibaba Qwen
Native 2K images with clean lettering
Qwen Research License(opens in a new tab)
Non-commercial use only
Qwen-Image 2.1 Turbo
Alibaba Qwen
8-step edits with up to 4 references, close to the full model except on big camera turns
Qwen Research License(opens in a new tab)
Non-commercial use only
Qwen-Image 2.1 · Small GPU
Alibaba Qwen
Qwen-Image 2.1 on a 16 GB GPU with a compact text reader
Qwen Research License(opens in a new tab)
Non-commercial use only
Qwen-Image 2.1 Live
Alibaba Qwen
Live edits as you type with the 2-step DMD2 LoRA
Qwen Research License(opens in a new tab)
Non-commercial use only
Tap to Recolor
Alibaba Qwen + Meta SAM 3.1
Only the tapped piece takes the new colour
Qwen Research License(opens in a new tab)
Non-commercial use only
FLUX 3 Klein
Black Forest Labs
BFL's open 5.6B model
FLUX Kommunity License v1.0(opens in a new tab)
Commercial use with conditions
FLUX 3 Klein Base
Black Forest Labs
The undistilled Klein your FLUX 3 Klein LoRAs are trained on
FLUX Kommunity License v1.0(opens in a new tab)
Commercial use with conditions
Krea 2 Realism
Krea
Natural skin, light and faces from detailed prose
Krea 2 Community License(opens in a new tab)
Commercial use with conditions
Krea 2 Turbo
Krea
Krea 2 in 8 steps
Krea 2 Community License(opens in a new tab)
Commercial use with conditions
MOSS-Image
OpenMOSS
Clean Chinese and English text, follows layout boxes
FIBO 2 Dev
Bria
Structured JSON prompts for exact control
CC BY-NC 4.0(opens in a new tab)
Non-commercial use only
FIBO 2 Edit Turbo
Bria
4-step edits with up to 4 references
CC BY-NC 4.0(opens in a new tab)
Non-commercial use only
Krea 2 Edit
Krea
Instruction edits that keep the picture
Krea 2 Community License(opens in a new tab)
Non-commercial use only
Krea 2 Identity Edit
Krea
Restages a person from one photo
Krea 2 Community License(opens in a new tab)
Non-commercial use only
Krea 2 Turbo · Image to image
Krea
Your picture redrawn in Krea 2's look at the strength you pick
Krea 2 Community License(opens in a new tab)
Commercial use with conditions
Krea 2 Inpaint
Krea
Paints the masked area or extends the canvas
Krea 2 Community License(opens in a new tab)
Commercial use with conditions
Z-Image Turbo
Alibaba Tongyi
Photoreal images in seconds
Qwen-Image 2512
Alibaba Qwen
Sharp lettering and strong composition
Qwen-Image Lightning
Alibaba Qwen
Qwen-Image, faster
FLUX.2 [dev]
Black Forest Labs
FLUX.2 with multi-reference editing
FLUX [dev] Non-Commercial License(opens in a new tab)
Non-commercial use only
FLUX.2 [dev] Turbo
Black Forest Labs
FLUX.2, faster
FLUX [dev] Non-Commercial License(opens in a new tab)
Non-commercial use only
FLUX.1 Krea [dev]
Black Forest Labs × Krea
Natural skin and light, true photorealism
FLUX.1 [dev] Non-Commercial License(opens in a new tab)
Non-commercial use only
FLUX.1 [dev]
Black Forest Labs
The classic FLUX, with thousands of styles
FLUX.1 [dev] Non-Commercial License(opens in a new tab)
Non-commercial use only
FLUX.1 [schnell]
Black Forest Labs
FLUX at its fastest
HiDream-I1 Dev
HiDream.ai
Follows long prompts closely
Llama 3.1 Community License(opens in a new tab)
Commercial use with conditions
Credit: Built with Llama
SDXL
Stability AI
The classic, with thousands of styles
CreativeML OpenRAIL++-M(opens in a new tab)
Commercial use with conditions
Qwen-Image-Edit 2511
Alibaba Qwen
Edit with words, up to 3 references
FLUX.1 Kontext [dev]
Black Forest Labs
Open instruction editing that preserves identity
FLUX.1 [dev] Non-Commercial License(opens in a new tab)
Non-commercial use only
FLUX.1 Fill [dev]
Black Forest Labs
Inpaint with a mask, or expand the canvas
FLUX.1 [dev] Non-Commercial License(opens in a new tab)
Non-commercial use only
ReActor face swap
Open source (inswapper)
Classic fast face swap with face restoration
InsightFace inswapper (research only)(opens in a new tab)
Non-commercial use only
4x ESRGAN upscale
Open source
Fast 4x image upscale
VTracer
visioncortex
Traces logos, icons and flat art into a clean, editable SVG
SeedVR2 (image)
ByteDance Seed
Diffusion upscaler that restores real detail
BiRefNet background removal
Open source
Clean cut-outs in one click
FeyNoBg fine hair
1038lab
Cut-outs that keep every strand of hair, fur and see-through edges
Not stated
From picture · first frame
Black Forest Labs × Openjutsu
Your character in the pose of the clip's first frame, in your picture's setting and light
FLUX Kommunity License v1.0(opens in a new tab)
Commercial use with conditions
Qwen-Image-Edit Relight
Alibaba Qwen
Relight a photo from a description (Relight LoRA, 4 steps)
Qwen-Image-Edit Relight (reference)
Alibaba Qwen
Relight a photo to match a reference image's light (Relight LoRA, 4 steps)
Qwen-Image 2.1 Next Scene
Alibaba Qwen
The next storyboard panel from the previous one, on Qwen-Image 2.1 with its Next-Scene LoRA
Qwen Research License(opens in a new tab)
Non-commercial use only
Relight
Alibaba Qwen
Relight with placed lights
Qwen Research License(opens in a new tab)
Non-commercial use only
Relight scene read
Open source
Depth and subject cut-out for the Relight preview
QMA · 96 views
Alibaba Qwen
Re-shoot a picture from any of 96 camera angles
Qwen Research License(opens in a new tab)
Non-commercial use only
Qwen-Image 2.1 Head Swap
Alibaba Qwen × Alissonerdx
Puts the whole head from image 2, hair included, on the person in image 1
Qwen Research License(opens in a new tab)
Non-commercial use only
Qwen-Image-Edit Next Scene
Alibaba Qwen
The next storyboard panel from the previous one (next-scene LoRA)
Qwen-Image 2.1 Restore
Alibaba Qwen × AusBoss
Old, worn or small photos back sharp on the same frame, then enlarged 2x (Restore LoRA + SeedVR2)
Qwen Research License(opens in a new tab)
Non-commercial use only
Qwen-Image-Layered
Alibaba Qwen
Splits a picture into editable transparent layers, back to front
Qwen-Image-Layered Control
DiffSynth-Studio
Pulls the element you name (“the guitar”) onto its own layer
Qwen-Image-Edit Style Reference
Alibaba Qwen
A new image in the look of your moodboard (1–2 collages)
FLUX.2 [klein] Style Reference
Black Forest Labs
Fast style transfer from up to 4 references, 4 steps
Krea 2 Style Reference
Krea
Trained for style reference (1–2 images)
Krea 2 Community License(opens in a new tab)
Commercial use with conditions
USO Style Reference
ByteDance
Style-only reference on FLUX.1 [dev]
FLUX.1 [dev] Non-Commercial License(opens in a new tab)
Non-commercial use only
Z-Image Turbo · Palette
Alibaba Tongyi
Generates in your hex colors and keeps them with a color transfer
Any Angle
Qwen (Alibaba) · lilylilith
The same photo from a camera you pick
Qwen Research License(opens in a new tab)
Non-commercial use only
SAM 3.1 person mask
Meta
A black-and-white mask of the person in a picture
Meta SAM License(opens in a new tab)
Commercial use with conditions
FLUX.2 [klein] Try-On
Black Forest Labs × xocialize
Dresses the person in image 1 in the garment in image 2 (a top, dress or outfit) and an optional bottom in image 3
Qwen-Image 2.1 Outfit Swap
Alibaba Qwen × AusBoss
Only the clothes change
Qwen Research License(opens in a new tab)
Non-commercial use only
Qwen-Image-Edit Product Shot
Alibaba Qwen × dx8152
Cuts your product out and relights it into the scene you describe (White_to_Scene LoRA, 4 steps)
Qwen-Image 2.1 Steam It
Alibaba Qwen
Finds the clothes and presses them smooth
Qwen Research License(opens in a new tab)
Non-commercial use only
Z-Image Turbo · Grid
Alibaba Tongyi
One prompt, a whole contact sheet (2×2 to 4×4) that shares one look, cut into tiles
Z-Image Turbo · Soul Cast sheet
Alibaba Tongyi
A cast member as a turnaround sheet (front, three-quarter, profile, back) plus the four views
Body Double
Naturalpad InstantHMR + Meta MHR
Every person's 3D body from a photo
Meta SAM License(opens in a new tab)
Non-commercial use only
Pixelate faces
Meta SAM 3.1 + ComfyUI
SAM 3.1 finds every face and covers it with large blocks
Meta SAM License(opens in a new tab)
Commercial use with conditions
Watermark Remover
Qwen-Image 2.1 + ComfyUI
Repairs only under the marks kept for removal
Qwen Research License(opens in a new tab)
Non-commercial use only
FLUX.2 [dev] Image Edit (1 reference)
black-forest-labs
A ready-made ComfyUI workflow: edit
FLUX [dev] Non-Commercial License(opens in a new tab)
Non-commercial use only
Qwen-Image-Edit 2509 (Lightning 4-step)
Comfy-Org
A ready-made ComfyUI workflow: edit
Qwen-Image-Edit 2511
Comfy-Org
A ready-made ComfyUI workflow: edit
Remove Background (ComfyUI-RMBG, BEN2)
A ready-made ComfyUI workflow: bg remove
Remove what was brushed, the rest unchanged (Qwen-Image 2.1, ComfyUI-Qwen-Watermark)
Comfy-Org
A ready-made ComfyUI workflow: watermark remove
Qwen Research License(opens in a new tab)
Non-commercial use only
Qwen-Image 2.1 Text to Image · 2 steps, DMD2 LoRA v1.5
Comfy-Org
A ready-made ComfyUI workflow: t2i
Qwen Research License(opens in a new tab)
Non-commercial use only
Qwen-Image 2.1 Text to Image · 4 steps, LightX2V Lightning LoRA
Comfy-Org
A ready-made ComfyUI workflow: t2i
Qwen Research License(opens in a new tab)
Non-commercial use only
Qwen-Image 2.1 Text to Image · 6 steps, P-Q-Image v0.2 upsampled LoRA (writes the details)
Comfy-Org
A ready-made ComfyUI workflow: t2i
Qwen Research License(opens in a new tab)
Non-commercial use only
Qwen-Image 2.1 Image Edit (up to 4 references) · 2 steps, DMD2 LoRA v1.5
Comfy-Org
A ready-made ComfyUI workflow: edit
Qwen Research License(opens in a new tab)
Non-commercial use only
Qwen-Image 2.1 Image Edit (up to 4 references) · 4 steps, LightX2V Lightning LoRA
Comfy-Org
A ready-made ComfyUI workflow: edit
Qwen Research License(opens in a new tab)
Non-commercial use only
Qwen-Image 2.1 Image Edit (up to 4 references) · 6 steps, P-Q-Image v0.2 upsampled LoRA (writes the details)
Comfy-Org
A ready-made ComfyUI workflow: edit
Qwen Research License(opens in a new tab)
Non-commercial use only
Qwen-Image 2.1 Inpaint · 2 steps, DMD2 LoRA v1.5
Comfy-Org
A ready-made ComfyUI workflow: inpaint
Qwen Research License(opens in a new tab)
Non-commercial use only
Qwen-Image 2.1 Inpaint · 4 steps, LightX2V Lightning LoRA
Comfy-Org
A ready-made ComfyUI workflow: inpaint
Qwen Research License(opens in a new tab)
Non-commercial use only
Qwen-Image 2.1 Expand · 2 steps, DMD2 LoRA v1.5
Comfy-Org
A ready-made ComfyUI workflow: outpaint
Qwen Research License(opens in a new tab)
Non-commercial use only
Qwen-Image 2.1 Expand · 4 steps, LightX2V Lightning LoRA
Comfy-Org
A ready-made ComfyUI workflow: outpaint
Qwen Research License(opens in a new tab)
Non-commercial use only
Qwen-Image 2.1 Outfit Swap (outfit in words; Outfit Swap LoRA)
Comfy-Org
A ready-made ComfyUI workflow: edit
Qwen Research License(opens in a new tab)
Non-commercial use only
Fuobe DoP
Fuobe Original
Camera motions and effects with their LoRAs on Wan 2.2
Fuobe DoP · LTX
Fuobe Original
FPV, wind and transitions on LTX-2.3 with its camera LoRAs
LTX-2 Community License(opens in a new tab)
Commercial use with conditions
Fuobe Animation
Fuobe Original
Keyframes in, the in-betweens drawn, one finished animation
Fuobe Cinema
Fuobe Original
Cinema Studio's director, with camera moves
Fuobe Genjutsu · Motion
Fuobe Original
Your character performs the clip's moves
Fuobe Genjutsu Lite · Motion
Fuobe Original
The quick tier
Fuobe Genjutsu · Swap
Fuobe Original
Swap a thing, a person or an outfit for your reference
Meta SAM License(opens in a new tab)
Commercial use with conditions
Fuobe Genjutsu Lite · Swap
Fuobe Original
The quick tier
Meta SAM License(opens in a new tab)
Commercial use with conditions
Fuobe Genjutsu · Restyle
Fuobe Original
Every frame takes a Fuobe look, motion and framing kept
Fuobe Evolution
Fuobe Original
Takes your clip further in a creative direction
Fuobe Character swap
Fuobe Original
The person in your clip becomes your character
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
Fuobe Mixed Media
Fuobe Original
Every frame drawn again in the look
Fuobe Shorts
Fuobe Original
Your footage in a new look, motion kept
Fuobe Speak
Fuobe Original
A portrait speaks your voice track, lips, face and head in sync, long takes
Fuobe Speak · Motion
Fuobe Original
Speech with the body and the scene moving
Fuobe Restore
Fuobe Original
Old, faded or black-and-white footage back in colour and sharp
LTX-2 Community License(opens in a new tab)
Commercial use with conditions
MiniMax H3
MiniMax
Video with its own sound, from words, a start frame, or start and end frames
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
LTX-2.5
Lightricks
Open video with native synchronized audio, up to 10s
LTX-2 Community License(opens in a new tab)
Commercial use with conditions
Wan 2.2 14B Lightning
Alibaba Wan
Wan 2.2, fast
Wan 2.2 14B
Alibaba Wan
Full-quality Wan 2.2 from words, an image, or start and end frames
MiniMax H3 · FastH3 V2
FastVideo
MiniMax H3 in 8 steps on a 16 GB GPU, from words, a start frame, or start and end frames
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
MiniMax H3 · FastH3 Trim
FastVideo
MiniMax H3 from words on an 8-12 GB GPU
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
MiniMax H3 · H3 Flash
Video Rebirth
MiniMax H3 in 6 steps with a natural, filmic look, from words, a start frame, or start and end frames
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
Wan 2.2 5B
Alibaba Wan
720p video on a single consumer GPU
LTX-2
Lightricks
Fast open video with audio
LTX-2 Community License(opens in a new tab)
Commercial use with conditions
Kandinsky 6.0
Kandinsky Lab
Kandinsky 6.0
CC BY-NC 4.0(opens in a new tab)
Non-commercial use only. Not licensed in the EU, UK and South Korea
HunyuanVideo 1.5
Tencent
Lightweight 8.3B model with strong motion
Tencent Hunyuan Community License(opens in a new tab)
Not licensed in the EU, UK and South Korea
Wan VACE
Alibaba Wan
Reference-to-video
Wan 2.2 Fun Camera
Alibaba Wan
Image-to-video with explicit camera control (pan, zoom, roll)
Wan Animate 2
Alibaba Wan
Make any character copy a reference performance
Wan 2.2 Fun Control
Alibaba Wan
Pose/depth-guided motion transfer from a driving video
InfiniteTalk
MeiGen
Talking avatar from one photo (native, ~6s)
InfiniteTalk Long
MeiGen
Long talking videos from one photo
Wan 2.2 S2V
Alibaba Wan
Speech-to-video with body motion, ~14s
SeedVR2 (video)
ByteDance Seed
One-step diffusion restoration and upscale, in passes so long clips fit
Open-H3 Regenerate
NVIDIA
Redraws a soft clip at 1080p, 2K or 4K
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
Kandinsky VSR
Kandinsky Lab
Kandinsky 6.0
ESRGAN video upscale
Open source
Fast frame-by-frame 4x upscale, in passes
RIFE interpolation
Open source
FPS Boost to 24–120 fps and slow motion
RTX Video Super Resolution
NVIDIA
Fast upscale up to 4x on NVIDIA RTX cards
NVIDIA Software License Agreement(opens in a new tab)
Commercial use with conditions
RTX Frame Generation
NVIDIA
FPS Boost to 24–120 fps and slow motion on RTX 40 series cards and newer
NVIDIA Software License Agreement(opens in a new tab)
Commercial use with conditions
RTX Video HDR
NVIDIA
SDR footage turned into 10-bit HDR by NVIDIA's learned HDR, on NVIDIA RTX cards
NVIDIA Software License Agreement(opens in a new tab)
Commercial use with conditions
MMAudio
hkchengrex
Sound effects for silent clips
CC BY-NC 4.0(opens in a new tab)
Non-commercial use only
Wan 2.2 Fun VACE Extend
Alibaba Wan
Continues a clip from its last 16 frames — same look and characters
Wan 2.2 Fun VACE Reframe
Alibaba Wan
Outpaints a clip to a new aspect ratio
Wan-Move
Alibaba Wan
Draw motion paths on the first frame — things move exactly along them
Bernini-R Relight
Bernini
Relight a video from a description of the new light
BiRefNet video matte
ZhengPeng7
Cut the subject out of every frame — a transparent WebM, or its alpha mask
SAM 3 video cutout
Meta
Keep just what you name (“the dog”, “the red car”) — a transparent WebM, or its alpha mask
Meta SAM License(opens in a new tab)
Commercial use with conditions
Bria VRMBG 3.0
Bria
Steady, flicker-free cut-outs that follow the subject frame to frame — the same model as the cloud's Bria matte
Bria VRMBG-3.0 licence(opens in a new tab)
Non-commercial use only
CorridorKey 2.0
Corridor Digital
Neural green-screen keying
Corridor Key Licence 2.0(opens in a new tab)
Commercial use with conditions
Depth Anything 3
ByteDance
Per-frame depth maps for compositing and 3D effects
Video Depth Anything (small)
ByteDance
Temporally stable, flicker-free depth
Video Depth Anything (large)
ByteDance
Steady, detailed depth for video
CC BY-NC 4.0(opens in a new tab)
Non-commercial use only
Wan 2.1 VACE Remove
Alibaba Wan
Name what to remove and it's gone, every other pixel untouched
Meta SAM License(opens in a new tab)
Commercial use with conditions
Wan 2.1 VACE Remove (painted)
Alibaba Wan
Paint what to remove on the first frame, and it's gone from the shot
Meta SAM License(opens in a new tab)
Commercial use with conditions
VOID Remove
Netflix
Removes the object and what it causes — shadows, reflections, things it moves
CogVideoX License(opens in a new tab)
Commercial use with conditions
Wan 2.1 VACE Swap
Alibaba Wan
Name what to swap and your reference takes its place, every other pixel untouched
Meta SAM License(opens in a new tab)
Commercial use with conditions
Wan 2.1 VACE Insert
Alibaba Wan
Paint where it goes on the first frame, and your reference stays there for the whole shot (a still or slow camera)
H3 Character Swap · Relay 2.0
MiniMax × Akatz Labs
Character swap with smoother long takes
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
Group swap
MiniMax × Openjutsu
Up to four characters, each taking over the person you name
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
Scene edit
MiniMax × Openjutsu
Change a thing, remove it or put the people somewhere new
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
Fuobe Reshoot
MiniMax × Viggle
Same shot
From picture
MiniMax × Openjutsu
The first frame you approved, moving exactly like the person in the clip
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
MiniMax H3
MiniMax × Openjutsu
Your character from one photo performs the clip's movement in the photo's own setting
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
Fix faces
MiniMax × Openjutsu
Redraws small or soft faces in a finished video so they match your character's photo
MiniMax H3 Community License(opens in a new tab)
Non-commercial use only. Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
H3 Person Removal · Relay 2.0
MiniMax × Akatz Labs
Name a person and they're removed, with the scene rebuilt behind them
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
SAM 3.1 green mask · H3 Person Removal
Meta × Akatz Labs
Who H3 Person Removal will take out
Meta SAM License(opens in a new tab)
Commercial use with conditions
H3 Character Swap · studio segment
MiniMax × Akatz Labs
One Character Swap segment, continued from the take before
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
SAM 3.1 person mask · H3 Relay
Meta × Akatz Labs
Who to swap in a shot
Meta SAM License(opens in a new tab)
Commercial use with conditions
Wan 2.2 Fun VACE Restyle
Alibaba Wan
Re-renders a ≤5 s clip in the look of its restyled first frame, keeping the motion (depth)
Wan 2.2 Fun VACE Blockout
Alibaba Wan
Renders a grey-box blockout from its depth and a first frame
Z-Image Mixed Media
Alibaba Tongyi
Stylises every frame (sketch, comic, paper…) at 4–24 fps, outlines kept
HunyuanVideo-Foley (video)
Tencent
48 kHz foley in sync with your whole clip (up to 30 s), laid under the untouched picture
Tencent Hunyuan Community License(opens in a new tab)
Not licensed in the EU, UK and South Korea
Whisper captions
OpenAI × yuvraj108c
Transcribes and burns word-by-word captions in one job, short clips (whole clip in RAM)
CC BY-NC-SA 4.0(opens in a new tab)
Non-commercial use only
InfiniteTalk Lip Re-sync
MeiGen
Re-syncs the lips (and head) of real footage to a new voice, up to 40 s per job
LatentSync 1.6
ByteDance
Re-syncs only the mouth of real footage to a new voice
CreativeML OpenRAIL++-M
Commercial use with conditions
Wan 2.1 + Uni3C path
Alibaba (Wan, DAMO Uni3C)
A still filmed along the camera path you fly in the 3D scene
Hy-Nomas
Tencent Hunyuan
A world you describe once, then steer turn by turn with words and camera moves
Apache-2.0(opens in a new tab)
Non-commercial use only
Body Double motion capture
Naturalpad InstantHMR + Meta MHR
Motion capture from a clip
Meta SAM License(opens in a new tab)
Non-commercial use only
Wan 2.2 14B Image to Video + LoRA (camera/VFX)
Comfy-Org
A ready-made ComfyUI workflow: vfx
MiniMax H3 Text to Video + Audio (20 steps)
Kijai
A ready-made ComfyUI workflow: t2v
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
MiniMax H3 Image to Video + Audio (20 steps)
Kijai
A ready-made ComfyUI workflow: i2v
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
MiniMax H3 Start + End Frame to Video + Audio (20 steps)
Kijai
A ready-made ComfyUI workflow: flf2v
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
MiniMax H3 360-degree orbit of a still + Audio (Orbit LoRA, 28 steps)
Kijai
A ready-made ComfyUI workflow: i2v
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
MiniMax H3 Text to Video + Audio (Pika H3 Flex, 6 steps)
Kijai
A ready-made ComfyUI workflow: t2v
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
MiniMax H3 Image to Video + Audio (Pika H3 Flex, 6 steps)
Kijai
A ready-made ComfyUI workflow: i2v
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
MiniMax H3 Start + End Frame to Video + Audio (Pika H3 Flex, 6 steps)
Kijai
A ready-made ComfyUI workflow: flf2v
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
Video Upscale / Restore (SeedVR2 3B int8)
Comfy-Org
A ready-made ComfyUI workflow: upscale video
Video Upscale (Real-ESRGAN 4x)
Comfy-Org
A ready-made ComfyUI workflow: upscale video
Frame Interpolation (FILM, native)
Comfy-Org
A ready-made ComfyUI workflow: interpolate
Frame Interpolation (RIFE 4.26, native)
Comfy-Org
A ready-made ComfyUI workflow: interpolate
Frame Interpolation (RIFE VFI custom pack)
A ready-made ComfyUI workflow: interpolate
LTX-2.3 22B distilled text to video + LoRAs
Comfy-Org
A ready-made ComfyUI workflow: t2v
LTX-2 Community License(opens in a new tab)
Commercial use with conditions
MiniMax H3 text to video at 768p that keeps its latent for a 2K finish (FL2VA + LightX2V Turbo 4-step v1.2, 6 steps er_sde; ComfyUI-Open-H3-Regenerate saves the latent)
Comfy-Org
A ready-made ComfyUI workflow: t2v
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
Finish a 768p MiniMax H3 draft in 2K from its saved latent: Open-H3 Regenerate 2K without the decode and re-encode (ComfyUI-Open-H3-Regenerate + LBH packs)
Comfy-Org
A ready-made ComfyUI workflow: upscale video
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
Group swap, one by one: exactly 2 characters, each person redrawn on a pass of their own (Openjutsu advanced_maxq)
Comfy-Org
A ready-made ComfyUI workflow: video group swap
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
From picture, the video on native MiniMax H3 (DMAD, 6 steps) from the approved first frame and the part's normals
Comfy-Org
A ready-made ComfyUI workflow: video from picture
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
Chatterbox Multilingual
Resemble AI
Clone any voice from a short sample, in 23 languages
IndexTTS-2
Bilibili
Expressive zero-shot cloning
bilibili IndexTTS-2 Model Use License
Commercial use with conditions
Qwen3-TTS
Alibaba Qwen
Design a voice from a description, or clone one from a short sample, in 10 languages
VoxCPM2
OpenBMB
Design or clone a voice in studio-quality 48 kHz, in 30 languages
AuK
Tencent
Change what was said in a recording, or how it was said, or speak a line in a voice
Qwen Research License(opens in a new tab)
Non-commercial use only
Chatterbox Dialogue
Resemble AI
Two speakers, each cloned from a sample
Chatterbox Voice Conversion
Resemble AI
Say it in anyone's voice — keeps your timing and emotion
ACE-Step 1.5
ACE Studio × StepFun
Open music generation with lyrics
MiniMax Music 3
MiniMax
Full songs with expressive vocals up to 5 minutes, from your lyrics and a style
MiniMax-Music3 Community License(opens in a new tab)
Commercial use with conditions
Credit: MiniMax-Music3
YuE2
Multimodal Art Projection
Songs sung from your lyrics, or a cover of a song you add with a new style and words
CC BY-NC 4.0(opens in a new tab)
Non-commercial use only
Stable Audio 3 Small SFX
Stability AI
Sound effects from a description in 8 steps
Stability AI Community License(opens in a new tab)
Commercial use with conditions
Stable Audio Open 1.0
Stability AI
Effects and loops up to 47 s
Stability AI Community License(opens in a new tab)
Commercial use with conditions
HunyuanVideo-Foley
Tencent
48 kHz foley from a description
Tencent Hunyuan Community License(opens in a new tab)
Not licensed in the EU, UK and South Korea
MMAudio (text to audio)
hkchengrex
Sound effects from text, best at about 8 s
CC BY-NC 4.0(opens in a new tab)
Non-commercial use only
ThinkSound
FunAudioLLM
Sound effects from a description, up to 30 s
MelBand RoFormer + Qwen3-ASR
Open source
Splits the voice from the music and effects, and transcribes the voice with timings
DBS + Qwen3-ASR
Open source
Splits the dialogue from the music and effects (the split runs on a CPU) and transcribes it with timings
DBS (dialogue, music, effects)
Open source
Splits a soundtrack into dialogue, music and effects
DBS remix (dialogue, music, effects at set levels)
Open source
Takes the music and effects out from under the dialogue, or keeps them quietly
RE-USE (clean up voice)
NVIDIA
Removes noise, echo, clipping and codec damage from speech, restores muffled or phone audio to full band
OpenMDW-1.1(opens in a new tab)
Keep the licence text and NVIDIA's notices when you share the weights
YuE2 3B: a cover of a song (its melody, new style and lyrics)
Comfy-Org
A ready-made ComfyUI workflow: music
CC BY-NC 4.0(opens in a new tab)
Non-commercial use only
Clean the dialogue of a soundtrack and keep its music and effects (DBS + NVIDIA RE-USE)
nvidia
A ready-made ComfyUI workflow: enhance voice
OpenMDW-1.1(opens in a new tab)
Keep the licence text and NVIDIA's notices when you share the weights
Pixal3D
Tencent ARC
A textured 3D model (.glb) that stays true to your photo (camera-aware), plus a turntable video
TRELLIS.2
Microsoft
A textured 3D model (.glb) from one product photo, plus a turntable video
Hunyuan3D 2.1
Tencent
Untextured 3D shape from a photo
Tencent Hunyuan 3D 2.1 Community License(opens in a new tab)
Not licensed in the EU, UK and South Korea
Hi3D 1.0
Hi3D (Sparc3D)
Refines a 3D model's geometry at 512, 1024 or 1536 from the photo it was made from
Hi3D Open License Agreement 1.0(opens in a new tab)
Commercial use with conditions
Credit: Powered by Hi3D 1.0
TRELLIS.2 Multi-view
Tencent ARC × Microsoft
One textured 3D model (.glb) from photos of 2–4 sides of an object, plus a turntable video
Qwen-Image 2.1 → TRELLIS.2
Alibaba Qwen × Microsoft
Describe one object and get a textured 3D model with a turntable
TripoSplat
Tripo (VAST)
A soft, photoreal 3D capture of the subject in one photo, plus an orbit video
Any Angle 3D view
Tripo (VAST)
A 3D view of the photo's subject for picking camera angles
Idol 3D
Speridlabs
A Gaussian splat (.spz) of one object from one picture, plus an orbit video
Idol 3D quick mesh
Speridlabs
A rough vertex-coloured .glb from one picture, no textures
Idol 3D Mesh
Speridlabs
A quick 3D model (.glb) of a person or product from one photo, plus an orbit clip
MoGe 2 scene
Microsoft
The whole photo as a textured 3D mesh to fly a camera through, plus its depth map
Image to a cleaned Gaussian splat of one object (Idol, ComfyUI-Idol crop and pruning) + an orbit video
VAST-AI
A ready-made ComfyUI workflow: splat
Four views of one object from one picture (Idol splat, rendered from its predicted camera: front, left, back, right)
VAST-AI
A ready-made ComfyUI workflow: splat
OvisOCR3
ATH-MaaS
Every line of text in an image, with its box
From picture · what happens
Qwen × Openjutsu
What happens in a part of a clip, written for the first frame you picked
Describe my character
Qwen × Openjutsu
Who the character in your picture is and what they wear, in one sentence for the swap
Qwen3-ASR
Alibaba Qwen
Timed captions from speech in 30 languages (auto-detect)
Qwen3.5 4B translate
Alibaba Qwen
Translates text or timed caption lines, timings kept
Gemma 4 speech translate
Speech straight to translated text, no timings (voice-over dubs)
Reference brief (Gemma 4)
Watches a reference clip and writes up its style
Find watermarks
Qwen3-VL + ComfyUI
Finds text, logos, stickers, mosaics and blocks printed over a photo, with their boxes
Shakker-Labs AWPortrait-Z
Shakker-Labs
A look for Z-Image
Danrisi Lenovo UltraReal (phone realism)
Danrisi
A look for Z-Image
Danrisi GrainScape UltraReal (film grain)
Danrisi
A look for Z-Image
suayptalha Z-Image-Turbo Realism
suayptalha
A look for Z-Image
Yuuuna Realistic Snapshot v5
Yuuuna
A look for Z-Image
rkppvc Z-Image Amateur Photo
rkppvc
A look for Z-Image
renderartist Technically Color Z
renderartist
A look for Z-Image
AIImageStudio Reversal Film Gravure v2.0
AIImageStudio
A look for Z-Image
Quorlen Sunbleached Photograph
Quorlen
A look for Z-Image
Quorlen Behind Reeded Glass
Quorlen
A look for Z-Image
none declared(opens in a new tab)
Non-commercial use only
artificialguybr Cinematic Film Still Redmond
artificialguybr
A look for Z-Image
artificialguybr Analog Redmond
artificialguybr
A look for Z-Image
artificialguybr Film Grain Redmond
artificialguybr
A look for Z-Image
artificialguybr Fish Eye Redmond
artificialguybr
A look for Z-Image
artificialguybr Double Exposure Redmond
artificialguybr
A look for Z-Image
artificialguybr Polaroid Redmond
artificialguybr
A look for Z-Image
artificialguybr Snapchat Selfie Redmond
artificialguybr
A look for Z-Image
artificialguybr Pixel Art Redmond
artificialguybr
A look for Z-Image
artificialguybr 360 View Redmond
artificialguybr
A look for Z-Image
artificialguybr Poster Redmond
artificialguybr
A look for Z-Image
artificialguybr Album Cover Redmond
artificialguybr
A look for Z-Image
AlamandaCathartica Digicam
AlamandaCathartica
A look for Z-Image
frankbarbs DARKROOM Noir
frankbarbs
A look for Z-Image
tarn59 80s Air Brush
tarn59
A look for Z-Image
renderartist Classic Painting Z
renderartist
A look for Z-Image
jarod2212 Rebel Essence Grunge
jarod2212
A look for Z-Image
jarod2212 Rebel Essence Bikers
jarod2212
A look for Z-Image
jarod2212 Rebel Essence Emo
jarod2212
A look for Z-Image
neph1 1980s Horror Movies
A look for Z-Image
neph1 1980s SciFi Movies
A look for Z-Image
Ttio2 Pencil Sketch
Ttio2
A look for Z-Image
V67 Y2K Style
A look for Z-Image
V67 Double Exposure
A look for Z-Image
V67 Bokeh
A look for Z-Image
V67 Dreamy Glow
A look for Z-Image
V67 Negative Space
A look for Z-Image
V67 Twins and Mirror
A look for Z-Image
V67 Golden Hour
A look for Z-Image
V67 Rain effect
A look for Z-Image
V67 Dark horror atmosphere
A look for Z-Image
V67 Cyberpunk
A look for Z-Image
V67 Fashionable, modest, well-tailored
A look for Z-Image
V67 Angels and Demons
A look for Z-Image
V67 Symbolic Surrealism
A look for Z-Image
V67 Layered paper diorama
A look for Z-Image
Movie Wide Angle Z-image
A look for Z-Image
Amateur Photos 90s 77
A look for Z-Image
Last Days of Film - Early 1990s
A look for Z-Image
vhsCommercial
A look for Z-Image
Soft Fuji Film Lighting
A look for Z-Image
Cinematic Retro Photography
A look for Z-Image
Z-Image Professional Photographer
A look for Z-Image
Portrait Prime FE85F14
A look for Z-Image
Helios 44-2
A look for Z-Image
AfterDark (lomo night film)
A look for Z-Image
Neon Noir Style
A look for Z-Image
Hard Flash
A look for Z-Image
Glitch Effect
A look for Z-Image
Glitchcore
A look for Z-Image
artsyblackandwhitephoto
A look for Z-Image
Cyber-Y2K
A look for Z-Image
Fairycore (World Morph)
A look for Z-Image
Chrome Enhancer
A look for Z-Image
Bingguo Nail Art
A look for Z-Image
Holding glass / woman with a drink
A look for Z-Image
XUPLX Unusual Photo Montage
A look for Z-Image
2030s fashion
A look for Z-Image
Camera Tilt-up Overhead
A camera move for Wan 2.2
Camera Tilt-up Overhead
A camera move for Wan 2.2
Camera Tilt-down Undershot V2
A camera move for Wan 2.2
Camera Tilt-down Undershot V2
A camera move for Wan 2.2
Shaky, handheld camera
A camera move for Wan 2.2
LuisaP Extreme Handheld Camera
A camera move for Wan 2.2
Headmounted camera / Headbob
A camera move for Wan 2.2
LuisaP Extreme Bodycam
A camera move for Wan 2.2
LuisaP Extreme Bodycam
A camera move for Wan 2.2
First Person POV Parkour
A camera move for Wan 2.2
First Person POV Parkour
A camera move for Wan 2.2
LuisaP Driving Crazy
A camera move for Wan 2.2
Flow Camera
A camera move for Wan 2.2
Circular shooting
A camera move for Wan 2.2
ostris Wan2.2 I2V orbit (already in the map)
ostris
A camera move for Wan 2.2
ostris Wan2.2 I2V orbit (low)
ostris
A camera move for Wan 2.2
chsengni LTX-2.3 FPV motion
chsengni
A camera move for LTX-2.3
TheBurgstall EarthZoomOut (LTX-2)
TheBurgstall
A camera move for LTX
Lightricks LTX-2.5 Slow-Motion-Control
Lightricks
A camera move for LTX-2.5
joyfox LTX-2.3 Transition
joyfox
A camera move for LTX-2.3
SOLRICKS LTX-2.5 BTS Movie Set
SOLRICKS
A camera move for LTX-2.5
HF card: LTX-2.x Community; civitai 2941912 lists RentCivit only(opens in a new tab)
Non-commercial use only
Wiggle Redmond
A camera move for Wan 2.2
civitai permissions: Image, RentCivit (no third-party hosting)(opens in a new tab)
Commercial use on your own GPU only
Wiggle Redmond
A camera move for Wan 2.2
civitai permissions: Image, RentCivit (no third-party hosting)(opens in a new tab)
Commercial use on your own GPU only
EXPLOSION Redmond
An effect for Wan 2.2
EXPLOSION Redmond
An effect for Wan 2.2
LuisaP NUKE
An effect for Wan 2.2
Wan Video Burst Effect
An effect for Wan 2.2
Wan Video Burst Effect
An effect for Wan 2.2
Wan 2.2 Low Frozen effects
An effect for Wan 2.2
Metalcification Redmond
An effect for Wan 2.2
Metalcification Redmond
An effect for Wan 2.2
LuisaP Glitch and Argue (soft)
An effect for Wan 2.2
LuisaP Glitch and Argue (soft)
An effect for Wan 2.2
LuisaP Fly Away
An effect for Wan 2.2
LuisaP Fly Away
An effect for Wan 2.2
Item decomposition
An effect for Wan 2.2
DigitalByte XPLDV Exploded View (LTX-2.5)
DigitalByte
An effect for LTX-2.5
Wan2.2 coin fountain (金币喷泉)
An effect for Wan 2.2
Wan2.2 coin fountain (金币喷泉)
An effect for Wan 2.2
Lana Wind Motion (LTX-2.3)
An effect for LTX-2.3
Car Jump Motion (LTX-2.5 distilled)
An effect for LTX-2.5
Cinematic Smash Cut
An effect for Wan 2.2
ThunderHero Redmond
An effect for Wan 2.2
ThunderHero Redmond
An effect for Wan 2.2
Remade-AI Fire (Wan 2.1 T2V)
Remade-AI
An effect for Wan 2.1
Qwen-Image-Edit-2509 Relight (start frame)
Comfy-Org
An effect for Qwen-Image
Lightricks LTX-2.3 IC-LoRA Relight (video)
Lightricks
An effect for LTX-2.3
Virtual Try-On for FLUX.2 [klein] 4B (xocialize)
xocialize
Part of a Fuobe recipe for FLUX.2
Next Scene for Qwen-Image-Edit 2509 (lovis93)
lovis93
Part of a Fuobe recipe for Qwen-Image
Multiple-angles for Qwen-Image-Edit 2509 (dx8152)
dx8152
Part of a Fuobe recipe for Qwen-Image
AnyAngle for Qwen-Image 2.1 (lilylilith)
lilylilith
Part of a Fuobe recipe for Qwen-Image 2.1
FusioniX for Wan 2.1 I2V 14B (vrgamedevgirl84)
vrgamedevgirl84
Part of a Fuobe recipe for Wan 2.1
Lightricks LTX Restore IC-LoRA (archive restoration and colourisation)
Lightricks
Part of a Fuobe recipe for LTX-2.3
Lightricks LTX-2.5 Detail Refine IC-LoRA (restore to 4K, second pass)
Lightricks
Part of a Fuobe recipe for LTX-2.5
DMD2 2-step for Qwen-Image 2.1, v1.5 (linoyts)
linoyts
Part of a Fuobe recipe for Qwen-Image 2.1
Qwen Research License(opens in a new tab)
Non-commercial use only
Lightning 4-step for Qwen-Image 2.1 (LightX2V)
lightx2v
Part of a Fuobe recipe for Qwen-Image 2.1
Not stated(opens in a new tab)
No licence stated: see its page
P-Q-Image 6-step upsampled for Qwen-Image 2.1, v0.2 (PrunaAI)
PrunaAI
Part of a Fuobe recipe for Qwen-Image 2.1
Qwen Research License(opens in a new tab)
Non-commercial use only
Next Scene for Qwen-Image 2.1 (akhaliq)
akhaliq
Part of a Fuobe recipe for Qwen-Image 2.1
Qwen Research License(opens in a new tab)
Non-commercial use only
QMA Camera Angles for Qwen-Image 2.1 (gribbliest)
gribbliest
Part of a Fuobe recipe for Qwen-Image 2.1
Not stated(opens in a new tab)
No licence stated: see its page
AniSora 3.2 for Wan 2.2 I2V A14B, high-noise expert (bilibili Index-AniSora, extracted by Kijai)
Kijai
Part of a Fuobe recipe for Wan 2.2
Lightx2v 4-step for Wan 2.2 I2V A14B, low-noise expert (Comfy-Org repack)
Comfy-Org
Part of a Fuobe recipe for Wan 2.2
Crane up
Remade-AI
A camera move for Wan 2.1
Crane down
Remade-AI
A camera move for Wan 2.1
Muscle
Remade-AI
An effect for Wan 2.1
Disintegration Redmond WAN2 I2V 14B
artificialguybr
An effect for Wan 2.2
Squish
Remade-AI
An effect for Wan 2.1
Cakeify
Remade-AI
An effect for Wan 2.1
Inflate
Remade-AI
An effect for Wan 2.1
Deflate
Remade-AI
An effect for Wan 2.1
Crush
Remade-AI
An effect for Wan 2.1
Rotate
Remade-AI
An effect for Wan 2.1
Gun Shooting
Remade-AI
An effect for Wan 2.1
Bride
Remade-AI
An effect for Wan 2.1
Baby
Remade-AI
An effect for Wan 2.1
Puppy
Remade-AI
An effect for Wan 2.1
Classy
Remade-AI
An effect for Wan 2.1
Snow White
Remade-AI
An effect for Wan 2.1
Disney Princess
Remade-AI
An effect for Wan 2.1
VIP
Remade-AI
An effect for Wan 2.1
Mona Lisa
Remade-AI
An effect for Wan 2.1
Princess
Remade-AI
An effect for Wan 2.1
Jungle
Remade-AI
An effect for Wan 2.1
Pirate Captain
Remade-AI
An effect for Wan 2.1
Samurai
Remade-AI
An effect for Wan 2.1
Zen
Remade-AI
An effect for Wan 2.1
Assassin
Remade-AI
An effect for Wan 2.1
Warrior
Remade-AI
An effect for Wan 2.1
Painting
Remade-AI
An effect for Wan 2.1
Fus Ro Dah
Remade-AI
An effect for Wan 2.1
Hug Jesus
Remade-AI
An effect for Wan 2.1
Super Saiyan
Remade-AI
An effect for Wan 2.1
Robot Face Reveal
Remade-AI
An effect for Wan 2.1
Explode
Remade-AI
An effect for Wan 2.1
Electrify
Remade-AI
An effect for Wan 2.1
Dolly Effect
Remade-AI
A camera move for Wan 2.1
Hulk Transformation
Remade-AI
An effect for Wan 2.1
Selfie With Younger Self
Remade-AI
An effect for Wan 2.1
angry face
Remade-AI
An effect for Wan 2.1
kissing
Remade-AI
A camera move for Wan 2.1
crying
Remade-AI
An effect for Wan 2.1
Laughing
Remade-AI
An effect for Wan 2.1
Cartoon Jaw Drop
Remade-AI
An effect for Wan 2.1
Jumpscare
Remade-AI
An effect for Wan 2.1
Hero run
Remade-AI
A camera move for Wan 2.1
Arc shot
Remade-AI
An effect for Wan 2.1
Crane over the head
Remade-AI
A camera move for Wan 2.1
matrix shot
Remade-AI
A camera move for Wan 2.1
Car chase
Remade-AI
A camera move for Wan 2.1
Crash zoom in
Remade-AI
A camera move for Wan 2.1
Crash zoom out
Remade-AI
A camera move for Wan 2.1
wan21 i2v dolly zoom lora
ostris
An effect for Wan 2.1
Motion Lora Camera Push In Wan 14B 720p I2V
lovis93
A camera move for Wan 2.1
SkinMorph Redmond WAN2 I2V 14B
artificialguybr
An effect for Wan 2.2
H3 Character Swap LoRA V1.5 (Akatz Labs)
akatz-ai
Part of a Fuobe recipe for MiniMax H3
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
H3 Character Swap LoRA V1 (Akatz Labs)
akatz-ai
Part of a Fuobe recipe for MiniMax H3
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
MiniMax H3 FL2V Turbo 8-step (lightx2v)
lightx2v
Part of a Fuobe recipe for MiniMax H3
H3 Person Remover V1 (Akatz Labs)
akatz-ai
Part of a Fuobe recipe for MiniMax H3
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
Pika H3 Flex FL2VA (6 steps; words, first and last frames)
Pika-Labs
Part of a Fuobe recipe for MiniMax H3
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
Credit: MiniMax H3 is licensed under the MiniMax H3 Community License Agreement, Copyright © 2026 MiniMax. All Rights Reserved. FL2VA and Ref2VA adapters Copyright © 2026 Mellis, Inc. These adapters are Model Derivatives of MiniMax H3 and are distributed under the MiniMax H3 Community License Agreement. Files modified from MiniMax H3 carry notices stating that they have been modified.
Pika H3 Flex Ref2VA (6 steps; images, clips and voice references)
Pika-Labs
Part of a Fuobe recipe for MiniMax H3
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
Credit: MiniMax H3 is licensed under the MiniMax H3 Community License Agreement, Copyright © 2026 MiniMax. All Rights Reserved. FL2VA and Ref2VA adapters Copyright © 2026 Mellis, Inc. These adapters are Model Derivatives of MiniMax H3 and are distributed under the MiniMax H3 Community License Agreement. Files modified from MiniMax H3 carry notices stating that they have been modified.
Open-H3-Regenerate-2K LoRA (NVIDIA)
Efficient-Large-Model
Part of a Fuobe recipe for MiniMax H3
MiniMax H3 Community License(opens in a new tab)
Not licensed in the US, the EU, the UK and South Korea
Credit: MiniMax H3
MiniMax H3 FL2V Turbo 4-step v1.2 (lightx2v)
lightx2v
Part of a Fuobe recipe for MiniMax H3
ComfyUI-WanVideoWrapper
kijai
Nodes for Fuobe Speak, InfiniteTalk Long and InfiniteTalk Lip Re-sync
ComfyUI-VideoHelperSuite
Kosinkadink
Video loading and saving
ComfyUI-KJNodes
kijai
Utility nodes (GrowMaskWithBlur, resizing)
ComfyUI-Frame-Interpolation
Fannovel16
RIFE and FILM frame interpolation
ComfyUI-RMBG
1038lab
Nodes for Fuobe Try-On, FeyNoBg fine hair, Bria VRMBG 3.0 and 2 more
ComfyUI-ReActor
Gourieff
Nodes for ReActor face swap
ComfyUI-MMAudio
kijai
Nodes for MMAudio and MMAudio (text to audio)
Not stated(opens in a new tab)
No licence stated: see its page
ComfyUI_Fill-ChatterBox
filliptm
Nodes for Chatterbox Multilingual, Chatterbox Dialogue and Chatterbox Voice Conversion
Not stated(opens in a new tab)
No licence stated: see its page
ComfyUI-IndexTTS2
snicolast
Nodes for IndexTTS-2
Not stated(opens in a new tab)
No licence stated: see its page
ComfyUI-Video-Depth-Anything
yuvraj108c
Nodes for Video Depth Anything (small) and Video Depth Anything (large)
Not stated(opens in a new tab)
No licence stated: see its page
ComfyUI-MelBandRoFormer
kijai
Nodes for MelBand RoFormer + Qwen3-ASR
Not stated(opens in a new tab)
No licence stated: see its page
ComfyUI-QwenASR
1038lab
Nodes for MelBand RoFormer + Qwen3-ASR, DBS + Qwen3-ASR and Qwen3-ASR
ComfyUI-HunyuanVideo-Foley
phazei
Nodes for HunyuanVideo-Foley and HunyuanVideo-Foley (video)
Not stated(opens in a new tab)
No licence stated: see its page
ComfyUI-ToSVG
Yanick112
Nodes for VTracer
ComfyUI-Whisper
yuvraj108c
Word-timed captions in one job
CC BY-NC-SA 4.0(opens in a new tab)
Non-commercial use only
ComfyUI-LatentSyncWrapper
ShmuelRonen
Nodes for LatentSync 1.6
Not stated(opens in a new tab)
No licence stated: see its page
ComfyUI-ThinkSound_Wrapper
ShmuelRonen
Nodes for ThinkSound
Not stated(opens in a new tab)
No licence stated: see its page
fuobe-live
workingprototype
Fuobe's own live edit nodes (fuobe_live.py, shipped with Fuobe)
ausboss-nodes
ausboss
Crop For Inpaint, Stitch Inpaint and Crop + Rotate + Pad for Qwen-Image 2.1's inpaint and expand
zen-image-edit-comfyui
recoilme
Zen Image Edit's adapter loader and text encode
ComfyUI-QwenTTS
1038lab
Qwen3-TTS voice design and voice clone nodes
ComfyUI-VoxCPM2
Saganaki22
VoxCPM2 text to speech (voice design) and voice clone nodes
ComfyUI-CorridorKey2
hashu786
CorridorKey 2.0 nodes, with Corridor's CorridorKey 2.0 code inside unmodified
Corridor Key Licence 2.0(opens in a new tab)
Commercial use with conditions
ComfyUI-AuK
Tencent-Hunyuan
Tencent AuK's own nodes (speech editing and text to speech)
kandinsky6
kandinskylab
Kandinsky 6.0 text and image to video with sound
kandinsky6-sr
kandinskylab
Kandinsky VSR video upscaling (2x, 2.25x, 4x)
ComfyUI-Qwen-Watermark
overflowofedge
The Watermark Remover's nodes
h3_character_swap
akatz-labs
MiniMax H3 character swap nodes (Akatz Labs)
h3-relay
akatz-ai
H3 Relay 2.0 (Sendy Software LLC and contributors)
ComfyUI-Viggle-Remake
Viggle
Viggle Remake's nodes (Viggle AI)
ComfyUI-MiniMaxH3-Context-Loop
ethanfel
MiniMax H3 Context Loop (NikoDemon80, Ethanfel)
ComfyUI-Fantastic-MiniMaxH3-PromptBuilder
Adudeguyman
The masked edit latent the H3 edits start from (source, mask and sound)
comfyui-h3-oneclick-swap
D3voz
D3voz's swap, scene edit and face section nodes (plans, role masks, grey mannequins)
comfyui-image-detail-contrast
D3voz
Softens fine detail contrast for Fix colours
Orion4D_FXMax
orion4d
Philippe Joye's DCTL tone mapper (Fix colours' film finish)
SolarRenderer
Enderfga
Nodes for From picture
ComfyUI-H3-FaceRefine
Carasibana
Face tracking, masks and paste-back for Fix faces (registry 1.1.1)
See its repository(opens in a new tab)
No licence stated: see its page
ComfyUI-Impact-Pack
ltdrdata
Its SAM loader (Fix faces' face masks
comfyui_controlnet_aux
Fannovel16
Metric3D normals that carry the clip's motion into From picture
ComfyUI-LTXVideo
Lightricks
Lightricks' LTX nodes (IC-LoRA loader and guide, per-step tiled latent fusion)
LTX-2 Community License(opens in a new tab)
Commercial use with conditions
ComfyUI-DBS
hiroki-abe-58
Dialogue, music and effects separation
ComfyUI-Idol
Nynxz
Idol's crop modes, predicted camera and splat pruning
ComfyUI-HyNomas
Nynxz
Hy-Nomas world generation on ComfyUI's own loaders
comfyui-krea2edit
lbouaraba
Krea 2 Identity Edit's nodes (Krea 2 Edit and Identity Edit)
ComfyUI-Krea2-Ostris-Edit
ostris
Ostris' Krea 2 edit nodes (Krea 2 Inpaint)
ComfyUI-MiniMax-H3-Turbo
Larryvrh
larryvrh's turbo LoRA loader and sampler for MiniMax H3 (8 steps)
plaguekind-nodes
PlagueKind
PlagueKind's H3 SLA attention (needs Triton) and AdaLN LoRA fix for MiniMax H3
ComfyUI-Open-H3-Regenerate
rdh9616
rdh9616's Open-H3 Regenerate nodes (NVIDIA's refine recipe, with NVIDIA's fast H3 VAE decoder vendored)
Comfyui_Minimax_h3_latent_Upscaler
LBH-123-AI
LBH-123-AI's H3 latent upscaler nodes (the Open-H3 Regenerate nodes call them by name)
Comfyui-Hi3d-T8
T8mars
T8star's Hi3D nodes (model loader, mesh loader, refine)
Commercial use with conditions
ComfyUI-RE-USE
workingprototype
Nodes for RE-USE (clean up voice) and Clean the dialogue of a soundtrack and keep its music and effects (DBS + NVIDIA RE-USE)
ComfyUI-InstantHMR
workingprototype
Fuobe's pinned form of jtydhr88's InstantHMR nodes (only the loader and the predictor)
ComfyUI-BodyDouble
workingprototype
Nodes for Body Double and Body Double motion capture
ComfyUI-HyperFlow-Flash
Video-Rebirth
Video Rebirth's HyperFlow H3 Flash nodes (GGUF loader and step schedule)
ComfyUI-GGUF
city96
city96's GGUF loaders and dequantizing ops
ComfyUI_Nvidia_VFX_Nodes
Comfy-Org
Comfy-Org's NVIDIA VFX nodes 0.2.1
viggle_turbo
Viggle
Viggle's ComfyUI nodes for its Qwen-Image 2.1 turbo LoRA (the loader that reads the LoRA's strength from its metadata, and the 6-step schedule)
Qwen Research License(opens in a new tab)
Non-commercial use only
Fuobe
This app, open source
ComfyUI
Comfy Org
Runs the open models on your GPU
FFmpeg
The FFmpeg developers
Joins, encodes and reads video
Openjutsu's workflow and Akatz Labs' app
Comfy-Org
The Character swap's ComfyUI graph
InstantHMR nano,
Mohamed Adjel, Kahina Chalabi and Vincent Bonnet (Naturalpad), run in the browser with ONNX Runtime Web
Body Double Live
MediaPipe Tasks and the selfie multiclass segmentation model, copyright Google LLC: each frame's full outline, clothes and hair included
Body Double Live
Trained on the Divide and Remaster (DnR) v2 dataset
Darius Petermann, Gordon Wichern, Zhong-Qiu Wang and Jonathan Le Roux
Dialogue, music and effects separation
Some of its clips carry their own licences
BS-RoFormer
Phil Wang
Dialogue, music and effects separation
Music-Source-Separation-Training
Roman Solovyev (ZFTurbo)
Dialogue, music and effects separation
TIGER / Look2Hear
Kai Li
Dialogue, music and effects separation
RE-USE speech enhancement model
NVIDIA Corporation
Clean up voice
Openjutsu
D3voz: the graphs Fuobe sends to your GPU, the settings code Fuobe ported from it, the SolarRenderer long-clip patch and the 17 light looks' wording
Genjutsu: Group swap, Scene edit, From picture, Fix faces and Fix colours
Adapted from MOSS-Image's prompt-enhancer skill and its layout reference (commit 2b46b1e), copyright OpenMOSS
The Layout Poster skill and Plan layout
Pika H3 Flex FL2VA and Ref2VA adapters, copyright © 2026 Mellis, Inc.
the Pika H3 Flex modes of MiniMax H3 on your GPU (Fast · Pika Flex, References, Talk from a voice clip)
@anthropic-ai/sdk
Anthropic
The official TypeScript library for the Anthropic API
@breezystack/lamejs
Alex Zhukov
Pure JavaScript MP3 Encoder
@fal-ai/client
fal-ai
The fal.ai client for JavaScript and TypeScript
@gltf-transform/core
Don McCurdy
glTF 2.0 SDK for JavaScript and TypeScript, on Web and Node.js
@gltf-transform/extensions
Don McCurdy
Adds extension support to @gltf-transform/core
@gltf-transform/functions
Don McCurdy
Functions for common glTF modifications, written using the core API
@google/model-viewer
Easily display interactive 3D models on the web and in AR!
@sparkjsdev/spark
sparkjsdev
An advanced 3D Gaussian Splatting renderer for THREE.js
@tanstack/react-query
tannerlinsley
Hooks for managing, caching and syncing asynchronous and remote data in React
@xyflow/react
xyflow
React Flow - A highly customizable React library for building node-based editors and interactive flow charts
better-sqlite3
Joshua Wise
The fastest and simplest library for SQLite in Node.js
class-variance-authority
Joe Bell
Class Variance Authority 🧬
cmdk
Paco
Part of the app
cn
shadcn-ui
Fast, small, compiled class-name merging for Tailwind CSS
lucide-react
Eric Fennis
A Lucide icon library package for React applications
meshoptimizer
Arseny Kapoulkine
Mesh optimization library that makes meshes smaller and faster to render
nanoid
Andrey Sitnik
A tiny (117 bytes), secure URL-friendly unique string ID generator
next
vercel
The React Framework
next-themes
pacocoursey
Part of the app
playwright-core
Microsoft Corporation
A high-level API to automate web browsers
radix-ui
radix-ui
Part of the app
react
react
React is a JavaScript library for building user interfaces
react-dom
react
React package for working with the DOM
react-markdown
Espen Hovlandsdal
React component to render markdown
remark-gfm
Titus Wormer
remark plugin to support GFM (autolink literals, footnotes, strikethrough, tables, tasklists)
server-only
This is a marker package to indicate that a module can only be used in Server Components
shadcn
shadcn
Add components to your apps
sharp
Lovell Fuller
High performance Node.js image processing, the fastest module to resize JPEG, PNG, WebP, GIF, AVIF and TIFF images
sonner
Emil Kowalski
An opinionated toast component for React
three
mrdoob
JavaScript 3D library
tw-animate-css
Luca Bosin
TailwindCSS v4.0 compatible replacement for `tailwindcss-animate`
unpdf
Johann Schopplich
PDF extraction and rendering across all JavaScript runtimes
zod
Colin McDonnell
TypeScript-first schema declaration and validation library with static type inference
This list is a guide, not legal advice. Read a model's own licence before you sell what you make with it.