jmtl/hk-cantonese-cc0-qwen3-tts-GGUF
Adapter vs jmtl/hk-cantonese-cc0-qwen3-tts. jmtl/hk-cantonese-cc0-qwen3-tts is a LoRA adapter, not full weights: rank 64, on down_proj, gate_proj, k_proj, o_proj, q_proj, up_proj, v_proj in 28 layers (196 adapted matrices). It modifies jmtl/hk-cantonese-cc0-qwen3-tts-GGUF at load time.