OsaurusAI/Raptor-0.6-4B-GGUF
Quantization vs XHToken/Spark-X2.5-4B. OsaurusAI/Raptor-0.6-4B-GGUF is a GGUF quantization (Q6_K) of the same weights: 73 tensors match exactly and 181 within quantization error.
Tensors changed: Embeddings 0/1, Norms 0/73, MLP 0/108, Attention 0/72