bosonai/higgs-audio-v3-stt-v2
LoRA-style vs bosonai/higgs-audio-v3-stt. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, k_proj, o_proj, out_proj, q_proj, up_proj, v_proj changed, in 32 of 32 layers; norms, biases and embeddings are untouched.
Tensors changed: other 0/10, Embeddings 0/3, Norms 0/243, MLP 84/212, Attention 240/336, lm_head 0/2
1 other repos hold byte-for-byte identical weight files.