kai-os/Carnice-V2-27b
LoRA-style vs Qwen/Qwen3.6-27B. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of in_proj_a, in_proj_b, in_proj_qkv, in_proj_z, k_proj, o_proj, out_proj, q_proj, v_proj changed, in 64 of 64 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/4, MLP 0/304, Attention 304/556, lm_head 0/1, Norms 0/319
7 derivatives checked
- kai-os/Carnice-V2-27b-GGUF: Same weights, converted
- bartowski/kai-os_Carnice-V2-27b-GGUF: Same weights, converted
- mradermacher/Carnice-V2-27b-i1-GGUF: Quantization
- mradermacher/Carnice-V2-27b-GGUF: Quantization
- massster/Carnice-V2-27b-GGUF: Same weights, converted
- stuchapin/Carnice-V2-27B-MTP-GGUF: Quantization
- sakamakismile/Carnice-V2-27b-NVFP4-TEXT-MTP: Quantization
1 other repos hold byte-for-byte identical weight files.