mradermacher/EVisRAG-3B-i1-GGUF
Quantization (fewer layers) vs Boggy666/EVisRAG-3B. mradermacher/EVisRAG-3B-i1-GGUF matches part of Boggy666/EVisRAG-3B (435 of its 825 tensors; the rest of Boggy666/EVisRAG-3B has no counterpart). On the shared part: mradermacher/EVisRAG-3B-i1-GGUF is a GGUF quantization (Q6_K) of the same weights: 181 tensors match exactly and 254 within quantization error.
Tensors changed: lm_head 0/1, MLP 0/108, Norms 0/73, Attention 0/252, Embeddings 0/1