nopesadly/higgs-audio-v3-8b-stt-v2-Q4_K_M-GGUF
Quantization (fewer layers) vs bosonai/higgs-audio-v3-8b-stt-v2. nopesadly/higgs-audio-v3-8b-stt-v2-Q4_K_M-GGUF matches part of bosonai/higgs-audio-v3-8b-stt-v2 (398 of its 894 tensors; the rest of bosonai/higgs-audio-v3-8b-stt-v2 has no counterpart). On the shared part: nopesadly/higgs-audio-v3-8b-stt-v2-Q4_K_M-GGUF is a GGUF quantization (Q4_K_M) of the same weights: 145 tensors match exactly and 253 within quantization error.
Tensors changed: Embeddings 0/1, Norms 0/145, MLP 0/108, Attention 0/144