spoonn/GLM-OCR-BF16-ComfyUI
Image-to-Text • Updated
Native ComfyUI model repackages
Converted weights with original licenses, provenance and links to native ComfyUI integration commits.
Note Model weights: MIT. Native core integration; documented OCR limitations.
spoonn/YingMusic-Singer-Plus-ComfyUI
Audio-to-Audio • UpdatedNote Main model: CC BY 4.0. Bundled VAE: Stability AI Community License; commercial conditions apply.
spoonn/AuK-Flash-ComfyUI
Text-to-Speech • UpdatedNote MIT generator/VAE plus Qwen RESEARCH LICENSE conditioning encoder: research/evaluation only; commercial permission required. Listening/edit-quality acceptance incomplete.
spoonn/Audio-Flamingo-3-BF16-ComfyUI
Audio-Text-to-Text • UpdatedNote NVIDIA OneWay Noncommercial License: research purposes only. Short-clip validation and limits documented.
spoonn/MOSS-Audio-8B-Thinking-ComfyUI
Audio-Text-to-Text • UpdatedNote Apache-2.0. Native BF16 package with retained reference/GPU/lifecycle evidence.
spoonn/MOSS-Audio-8B-Instruct-ComfyUI
Audio-Text-to-Text • UpdatedNote Apache-2.0. Native BF16 package with provenance; variant-specific acceptance remains limited.
spoonn/SoulX-Singer-Native-ComfyUI
Audio-to-Audio • UpdatedNote Main models: Apache-2.0. Dereverb: GPL-3.0 with source included. Optional karaoke separator withheld: no declared original license. Song-quality acceptance incomplete.
spoonn/Qwen3-Omni-30B-A3B-NVFP4-ComfyUI
Image-Text-to-Text • UpdatedNote Apache-2.0. Native Thinker text/image/audio understanding; source-model failures and NVFP4 runtime differences documented.