OmniRT supported models
This page is generated from the live registry by scripts/generate_models_doc.py.
Update the registry, not this file.
Models are organized by the digital-human production chain rather than by a generic multimodal taxonomy.
Core avatar rendering
Registry id
Task
Tier
Maturity
Realtime
Summary
longcat-video-avatar-1.5audio2videocore
beta
no
LongCat-Video-Avatar 1.5 audio-driven avatar generation via external CUDA or Ascend checkouts.
soulx-flashhead-1.3baudio2videocore
beta
no
SoulX-FlashHead low-latency talking-head generation via image plus audio.
soulx-flashtalk-14baudio2videocore
beta
yes
SoulX-FlashTalk talking-head avatar generation via image plus audio on CUDA or Ascend.
soulx-liveact-14baudio2videocore
beta
no
SoulX-LiveAct long-form audio-driven avatar video generation on Ascend.
Voice generation
Registry id
Task
Tier
Maturity
Realtime
Summary
cosyvoice3-triton-trtllmtext2audiocore
beta
no
CosyVoice3 text-to-audio generation through a Triton-compatible service endpoint.
indexttstext2audioadjacent
beta
yes
IndexTTS-2 resident text-to-audio service for OpenTalking TTS, with segment streaming and experimental token-window streaming through serve-text2audio.
soulx-podcast-1.7btext2audiocore
beta
no
SoulX-Podcast text-to-audio generation through a FastAPI service endpoint.
Voice understanding roadmap
Registry id
Task
Tier
Maturity
Realtime
Summary
sensevoice-smallaudio2textcore
beta
no
SenseVoice offline audio transcription for digital-human voice understanding.
Avatar asset generation
Registry id
Task
Tier
Maturity
Realtime
Summary
chronoediteditexperimental
beta
no
ChronoEdit physically-consistent image editing pipeline.
flux-cannyeditadjacent
beta
no
Flux canny-guided structured image generation pipeline.
flux-deptheditadjacent
beta
no
Flux depth-guided structured image generation pipeline.
flux-kontexteditadjacent
beta
no
Flux Kontext image editing pipeline.
qwen-image-editeditadjacent
beta
no
Qwen-Image single-image editing pipeline.
qwen-image-edit-pluseditadjacent
beta
no
Qwen-Image multi-reference editing pipeline.
qwen-image-layerededitadjacent
beta
no
Qwen-Image layered decomposition pipeline.
sdxl-refiner-1.0image2imageadjacent
beta
no
SDXL refiner image-to-image pipeline for second-stage refinement passes.
flux-fillinpaintadjacent
beta
no
Flux Fill inpainting and outpainting pipeline.
bria-3.2text2imageexperimental
beta
no
Bria 3.2 commercial-ready text-to-image pipeline.
flux-devtext2imageadjacent
stable
no
Flux 1 dev text-to-image pipeline.
flux-schnelltext2imageexperimental
stable
no
Flux 1 schnell low-step text-to-image pipeline.
flux2.devtext2imageadjacent
beta
no
Flux 2 dev text-to-image pipeline.
glm-imagetext2imageexperimental
beta
no
GLM-Image instruction-following text-to-image pipeline.
hidream-i1text2imageexperimental
beta
no
HiDream-I1 modern text-to-image pipeline.
hunyuan-image-2.1text2imageexperimental
beta
no
Hunyuan Image 2.1 text-to-image pipeline.
kolorstext2imageexperimental
beta
no
Kolors multilingual text-to-image pipeline.
lumina-t2xtext2imageexperimental
beta
no
Lumina-T2X text-to-image pipeline via the LuminaPipeline runtime.
omnigentext2imageexperimental
beta
no
OmniGen text-to-image generation path.
ovis-imagetext2imageexperimental
beta
no
Ovis-Image text-heavy generation pipeline.
pixart-sigmatext2imageexperimental
beta
no
PixArt-Sigma high-resolution text-to-image pipeline.
qwen-imagetext2imageadjacent
beta
no
Qwen-Image multilingual text-to-image pipeline.
sana-1.6btext2imageexperimental
beta
no
Sana 1.6B efficient text-to-image pipeline.
sd15text2imageexperimental
beta
no
Stable Diffusion 1.5 baseline text-to-image pipeline.
sd21text2imageexperimental
beta
no
Stable Diffusion 2.1 text-to-image pipeline.
sd3-mediumtext2imageexperimental
beta
no
Stable Diffusion 3 Medium text-to-image pipeline.
sd3.5-largetext2imageexperimental
beta
no
Stable Diffusion 3.5 Large text-to-image pipeline.
sd3.5-large-turbotext2imageexperimental
beta
no
Stable Diffusion 3.5 Large Turbo text-to-image pipeline.
sdxl-base-1.0text2imageadjacent
stable
no
SDXL base text-to-image pipeline with LoRA support.
sdxl-turbotext2imageexperimental
beta
no
SDXL Turbo low-latency text-to-image pipeline.
Video and idle assets
Registry id
Task
Tier
Maturity
Realtime
Summary
helios-i2vimage2videoexperimental
beta
no
Helios image-to-video pipeline.
hunyuan-video-1.5-i2vimage2videoexperimental
beta
no
HunyuanVideo 1.5 image-to-video pipeline.
kandinsky5-i2vimage2videoexperimental
beta
no
Kandinsky 5 Pro image-to-video pipeline.
ltx2-i2vimage2videoexperimental
beta
no
LTX image-to-video pipeline.
svdimage2videoadjacent
stable
no
Stable Video Diffusion base image-to-video pipeline.
svd-xtimage2videoadjacent
stable
no
Stable Video Diffusion XT image-to-video pipeline.
wan2.1-i2v-14bimage2videoexperimental
beta
no
Wan 2.1 image-to-video pipeline.
wan2.2-i2v-14bimage2videoadjacent
beta
no
Wan 2.2 image-to-video pipeline.
animate-diff-sdxltext2videoadjacent
beta
no
AnimateDiff SDXL text-to-video pipeline.
cogvideox-2btext2videoexperimental
beta
no
CogVideoX 2B text-to-video pipeline.
cogvideox-5btext2videoexperimental
beta
no
CogVideoX 5B text-to-video pipeline.
helios-t2vtext2videoexperimental
beta
no
Helios text-to-video pipeline.
hunyuan-videotext2videoexperimental
beta
no
HunyuanVideo text-to-video pipeline.
hunyuan-video-1.5-t2vtext2videoexperimental
beta
no
HunyuanVideo 1.5 text-to-video pipeline.
kandinsky5-t2vtext2videoexperimental
beta
no
Kandinsky 5 Pro text-to-video pipeline.
ltx-videotext2videoexperimental
beta
no
LTX-Video text-to-video pipeline.
mochitext2videoexperimental
beta
no
Mochi text-to-video pipeline.
sana-videotext2videoexperimental
beta
no
Sana-Video efficient text-to-video pipeline.
skyreels-v2text2videoexperimental
beta
no
SkyReels-V2 text-to-video pipeline.
wan2.1-t2v-14btext2videoexperimental
beta
no
Wan 2.1 text-to-video pipeline.
wan2.2-t2v-14btext2videoadjacent
beta
no
Wan 2.2 text-to-video pipeline.
Aliases
Alias
Canonical id
flux2-devflux2.dev