mstar.model#

Modules

`audio_watermark`	Opt-in AudioSeal watermark for generated speech.
`canary`
`canary_qwen`
`chatterbox`
`command_a_plus`
`cosmos3`	Cosmos3 omni generator model package.
`cosyvoice3`
`flux2_klein`	FLUX.2 [klein] (4B / 9B): distilled rectified-flow text-to-image and image editing.
`glm52`
`higgs_audio`
`kokoro`
`lingbot_world`
`ltx2_5`
`media_limits`	Decoded-size caps for request media.
`multimodal`	Order-preserving multimodal prompt adapter.
`nemotron_duplex`	NVIDIA NemotronLabs VoiceChat-11B (full-duplex speech-to-speech) support for M*.
`omnilingual_asr`
`omnivoice`
`orpheus`
`parakeet`
`pi05`
`qwen3_5`
`qwen3_asr`
`submodule_base`
`utils`
`vjepa2`
`wan22`
`waypoint`
`whisper`
`z_image`	Z-Image-Turbo: an 8-step distilled single-stream flow DiT (Qwen3-4B captions, FLUX.1 VAE).