shahvandit/qwen3-tts-tokenizer-multi-codebook-hf

Partial fine-tune vs Qwen/Qwen3-TTS-Tokenizer-12Hz. Partial change: 0% of the weights changed (other), in 16 of 31 layers.

Tensors changed: other 16/239, Norms 0/53, MLP 0/56, Attention 0/80, Embeddings 0/32

On the Hub · Downloads on Model Pulse