Model distribution — NOTICE

This Stability AI Model is licensed under the Stability AI Community License, Copyright © Stability AI Ltd. All Rights Reserved

Gemma is provided under and subject to the Gemma Terms of Use found at ai.google.dev/gemma/terms

MODIFICATIONS

NEIROA includes converted model files derived from the MLX checkpoints distributed in stabilityai/stable-audio-3-optimized.

- t5gemma_f16.npz was converted without tensor-value changes to t5gemma_f16.safetensors.
- TOKENIZER_MODEL was extracted from t5gemma_f16.npz as t5gemma_tokenizer.model.
- dit_sm-music_f16.npz was converted without tensor-value changes to dit_sm-music_f16.safetensors.
- cond.padding_embedding, cond.seconds_total_weight, and cond.seconds_total_bias were extracted from dit_sm-music_f16.npz as sa3_conditioner_sm-music.safetensors.
- same_s_decoder_f32.npz was converted without tensor-value changes to same_s_decoder_f32.safetensors.

The container format was changed from NumPy NPZ to Safetensors for local MLX Swift loading on iOS. The application uses the Stable Audio 3 Small Music variant and SAME-S decoder.

Powered by Stability AI