mradermacher/Qwen3-TTS-12Hz-1.7B-Vi-i1-GGUF
Quantization (fewer layers) vs beyoru/Qwen3-TTS-12Hz-1.7B-Vi. mradermacher/Qwen3-TTS-12Hz-1.7B-Vi-i1-GGUF matches part of beyoru/Qwen3-TTS-12Hz-1.7B-Vi (309 of its 480 tensors; the rest of beyoru/Qwen3-TTS-12Hz-1.7B-Vi has no counterpart). On the shared part: mradermacher/Qwen3-TTS-12Hz-1.7B-Vi-i1-GGUF is a GGUF quantization (Q6_K) of the same weights: 113 tensors match exactly and 196 within quantization error.
Tensors changed: Norms 0/113, MLP 0/84, Attention 0/112