aufklarer/Qwen3-TTS-12Hz-1.7B-Base-MLX-bf16

Same weights, converted vs Qwen/Qwen3-TTS-12Hz-1.7B-Base. Same weights stored in another dtype (BF16 → F16).

Tensors changed: other 0/79, lm_head 0/15, Embeddings 0/17, Norms 0/134, MLP 0/103, Attention 0/132

1 other repos hold byte-for-byte identical weight files.

On the Hub · Downloads on Model Pulse