Skip to content
EMPRESSION AI

Library

Models

A visual library of engines available to the studio.

Development Mode · labeled preview until engines are connected

Image

Qwen-Image

Text to image on weights you host.

Best for: Stills, edits, upscale, restore

Advanced details

License Apache-2.0

Availability · Preview adapter

Image

Qwen-Image-Edit

Edit an existing still with a prompt.

Best for: Stills, edits, upscale, restore

Advanced details

License Apache-2.0 · reference image

Availability · Preview adapter

Image

Real-ESRGAN

Upscale a still.

Best for: Stills, edits, upscale, restore

Advanced details

License BSD-3 · reference image

Availability · Preview adapter

Image

CodeFormer

Restore faces in a still.

Best for: Stills, edits, upscale, restore

Advanced details

License Non-commercial research — review before selling · reference image

Availability · Preview adapter

Image

BiRefNet

Remove the background.

Best for: Stills, edits, upscale, restore

Advanced details

License MIT · reference image

Availability · Preview adapter

Video

LTX-Video

Fast short clips from text or a still.

Best for: Short clips from text or a still

Advanced details

License Open weights

Availability · Preview adapter

Video

Wan 2.2

Open video quality leader, Apache-2.0.

Best for: Short clips from text or a still

Advanced details

License Apache-2.0

Availability · Preview adapter

Video

HunyuanVideo 1.5

Cinematic motion, FP8 on a 4090.

Best for: Short clips from text or a still

Advanced details

License Tencent community — read the card

Availability · Preview adapter

Video

CogVideoX-5B

Prompt-faithful short clips.

Best for: Short clips from text or a still

Advanced details

License Apache-2.0

Availability · Preview adapter

Lip-sync

MuseTalk

Replace the mouth on footage to match audio.

Best for: Match a mouth to audio you own

Advanced details

License MIT · reference image-or-video

Availability · Preview adapter

Lip-sync

LatentSync 1.6

Sharper 512px mouth replace.

Best for: Match a mouth to audio you own

Advanced details

License Apache-2.0 · reference image-or-video

Availability · Preview adapter

Lip-sync

Hallo2

Still photo plus audio becomes a talking person.

Best for: Match a mouth to audio you own

Advanced details

License MIT · reference image

Availability · Preview adapter

Lip-sync

LivePortrait

Animate a still from a driving video.

Best for: Match a mouth to audio you own

Advanced details

License MIT · reference image

Availability · Preview adapter

Lip-sync

InfiniteTalk

Longer talking video on Wan weights.

Best for: Match a mouth to audio you own

Advanced details

License Check checkpoint card · reference image-or-video

Availability · Preview adapter

Voice

Qwen3-TTS

Text to speech with a 3-second clone.

Best for: Speech from a script, or convert a voice

Advanced details

License Apache-2.0

Availability · Preview adapter

Voice

CosyVoice 3

Streaming TTS and clone.

Best for: Speech from a script, or convert a voice

Advanced details

License Apache-2.0

Availability · Preview adapter

Voice

Chatterbox

Fast MIT-licensed speech.

Best for: Speech from a script, or convert a voice

Advanced details

License MIT

Availability · Preview adapter

Voice

Kokoro

Tiny English TTS.

Best for: Speech from a script, or convert a voice

Advanced details

License Apache-2.0

Availability · Preview adapter

Voice

RVC

Convert one voice into another.

Best for: Speech from a script, or convert a voice

Advanced details

License MIT · reference audio

Availability · Preview adapter

Music

ACE-Step 1.5

Full songs with vocals from text.

Best for: Songs and beds on ACE-Step

Advanced details

License MIT / Apache

Availability · Preview adapter

Sound

Stable Audio Open

Short SFX and textures. Read the revenue cap.

Best for: SFX, textures, stems

Advanced details

License Stability community

Availability · Preview adapter

Sound

Demucs

Split vocals, drums, bass, other.

Best for: SFX, textures, stems

Advanced details

License MIT · reference audio

Availability · Preview adapter

Captions

faster-whisper

Speech to text for captions and dubbing.

Best for: Transcribe audio you upload

Advanced details

License MIT · reference audio

Availability · Preview adapter

Building on the API? See Developers.

Your Imagination Is Ready.

Enter Empression AI and begin creating across every medium.

Enter the Studio